University of Florida Thesis Or Dissertation Formatting

CHARACTERIZATION OF VOLATILE PYRETHROIDS FOR MOSQUITO MANAGEMENT

CHRISTOPHER STEPHEN BIBBS

A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA

2019

To my children, Aerith Liesel Bibbs and Griffin Roderick Bibbs, for being the reason I seek to better myself.

To my parents, for enabling my fascination with insects.

To my advisors, from my first days as an undergraduate to my last days completing my doctorate, whose friendship changed the way I think about the world.

And to all the wonderful people who saw fit to “pay it forward” throughout my journey, even though I was owed nothing.

ACKNOWLEDGMENTS

I greatly appreciate the Anastasia Mosquito Control District (AMCD) of St. Johns

County for allowing me to pursue my education with their employee degree work-study support.

If it were not for the policies enacted by the AMCD board of commissioners, the high-quality facilities opened to my research, and most importantly, the strong belief in higher education emphasized by District director Dr. Rui-De Xue, none of this would have been possible. I originally rejected the idea of going to school after my Master of Science. Dr. Xue did not give up on my potential to go farther. Despite the struggles, I am better for it. I hope the work enclosed in this dissertation has helped improve the visibility of AMCD as a profound public health research entity in Florida. Various staff and friends have all walked with me on this path.

A special debt of gratitude is owed to former biologist Alice Fulcher, former Education

Specialist Jodi Scott-Fiorenzano, Operations Manager Kay Gaines, Data Manager/Business

Manager Richard Weaver, former Field Biologist Michael Smith, senior Mechanic James Wynn, former Biological Technician Jennifer Gibson-Corrado, former head of surveillance Dr. Daniel

Dixon, Assistant Supervisor Dena Autry, Education Specialist Molly Clark, as well as all the

Mosquito Control Technicians, Seasonal Inspector/Sprayers, Visiting Scientists, and Interns.

These people have enriched my life, offered their guidance, and taught me public health as a passion and a career.

Funding for this research was leveraged in large part from the grants awarded by the

Florida Department of Agriculture and Consumer Services, with special mention to National

Science Foundation’s Center for Arthropod Management Technology (NSF-CAMTech), whose

Industry Advisory Board helped me on my path. When I started this rocky road, I lacked the experience, the notoriety, and the skills to receive even a single grant despite years of attempts prior to my PhD program. Aside from my advisor, Dr. Phillip Kaufman, whom I learned from

long before even graduating with my Bachlelor of Science, the contributions of Dr. Jeff

Bloomquist, Dr. Daniel Hahn, and Dr. Christopher Batich cannot be understated. Some believed that I may have chosen some of the most difficult committee members possible out of my options at the Entomology and Nematology Department. In my belief, if I was not good enough for them then I was not good enough for myself. We wrote these grant proposals together. We received this funding together. This dissertation was published by all of us together. Your experience, values, and fortitude are part of my own now.

My sponsorship as a student has been a joint effort by the University of Florida

Entomology and Nematology Department (UF), Florida Mosquito Control Association (FMCA),

American Mosquito Control Association (AMCA), the Florida Entomological Society (FES), and the Entomological Society of America (ESA). The entomology department at UF, through

Dr. Heather McAuslane, awarded the Grinter Fellowship three years to my PhD program. This great honor provided stability as I sought additional funding to complete my program. The

FMCA memorial scholarships of Cy Lesser and T. Wainwright, bestowed graciously by public health role models such as Dr. Roxanne Connelly, Chris Lesser of Manatee County Mosquito

Control District, Aaron Lloyd of Pasco County Mosquito Control District, and many others was a powerful support system that ensured I could make it to the finish line. The travel grants, student competition awards, and general student support networks of AMCA, FES, and ESA continuously proved how much the membership base was looking out for me. Dr.’s Ary Faraji,

Isik Unlu, Seth Britch, Mike Breidenbaugh, Michael Turrell, Stan Cope, Kristy Burkhalter,

Rajeev Vaidyanathan, Mustapha Debboun, Heather McAuslane, Blair Siegfried, Jawwad

Qureshi, and many other members have all been a wonderful part of my life as they have demonstrated how to be a good entomologists, even better scientists, and great human beings.

This series of projects was made possible by collaborations through the United States

Department of Agriculture – Agricultural Research Service – Center for Medical, Agricultural, and Veterinary entomology, McLoughlin Gormley King Corporation (MGK), Sumitomo

Corporation, and Sigma Scientific, LLC. Dr.’s Dan Kline, Bradley Willenberg, Maia Tsikolia,

Uli Bernier, Nurhayat Tabanca and Ken Linthicum have all been wonderful partners as we learn how to advance vector management. Dr. Jennifer Williams, Hitoshi Kawada, Takao Ishiwatari, and others affiliated with MGK and Sumitomo have been insightful collaborators that provided the tools that were needed to advance our understanding of my research topics. And finally,

Rudolph Strohschein and Jim Estaver have been inspirational creatives whose ingenuity and deep history with scientific process will be a lasting influence in my career. They may declare themselves just two hicks in Micanopy, FL, but Sigma will always be a beautiful cover for such diamonds in the rough.

To my advisors through time, Dr. Phillip Kaufman, Dr. Dawn Gouge, and Dr. Rebecca

Baldwin: this dissertation is proof your faith was not misplaced. I met all three of you when I was only an undergraduate at the University of Florida. I am glad that I have been able to keep you in my life and learn from you both before and after my various stepping stones. My very first attempt at being an entomologist was only successful because Dr. Baldwin rescued me, mentored me, and looked out for me as I navigated college life and my early learning as a researcher. When no other graduate program I applied for would give me a chance, Dr. Gouge offered a place in her piece of Arizona because she believed I could succeed. I hope I can give back to you even now that I have left your laboratory. Since the beginning, my greatest professional role model has always been Dr. Phillip Kaufman. When I did not know what to do, I would think back to my crowded, tiny work spaces shared with Dr. Dale Halbritter and Dr. Chris

Holderman and ask myself what Kaufman would say I needed to do. I emulated you and your values from the beginning. When I did what I thought Dr. Kaufman would tell me, I aced my medical, veterinary, biosecurity, and forensic entomology subjects. When I did what I thought

Dr. Kaufman would tell me, I completed my Master of Science at University of Arizona on time, with publications in print, and a job offer at AMCD waiting for me out the door. When I did what I thought Dr. Kaufman needed me to do, I accomplished the work in my PhD program despite being a non-traditional student, working a full-time job, and not having the benefit of being surrounded by peer students at a University office every day. I was always part of Dr.

Kaufman’s laboratory, even if I did not occupy a desk. Thank you for being there.

TABLE OF CONTENTS

page

ACKNOWLEDGMENTS ...... 4

LIST OF TABLES ...... 11

LIST OF FIGURES ...... 12

LIST OF ABBREVIATIONS...... 15

ABSTRACT ...... 18

CHAPTER

1 INTRODUCTION...... 20

1.1 Preface ...... 20 1.2 Literature Review ...... 24 1.2.1 Repellency ...... 24 1.2.2 Acute Symptoms and Toxicity ...... 29 1.2.3 Sub-Lethal Effects ...... 33 1.2.4 Evaluation Methods ...... 38 1.3 Objectives ...... 42 1.4 Discolsures ...... 43 1.5 Figures ...... 44

2 VAPOR TOXICITY OF FIVE VOLATILE PYRETHROIDS AGAINST FOUR MOSQUITO VECTOR SPECIES ...... 45

2.1 Introduction ...... 45 2.2 Materials and Methods ...... 47 2.2.1 Insect Strains ...... 47 2.2.2 Chemicals ...... 48 2.2.3 Fumigant Bioassay Design...... 49 2.2.4 Data Analysis ...... 50 2.3 Results ...... 51 2.4 Discussion ...... 55 2.5 Disclosures ...... 57 2.6 Tables...... 58 2.7 Figures ...... 62

3 SUB-LETHAL EFFECTS OF TRANSFLUTHRIN ON DOMESTIC MOSQUITO FECUNDITY AND OVIPOSITION BEHAVIOR ...... 65

3.1 Introduction ...... 65 3.2 Materials and Methods ...... 68

3.2.1 Insect Rearing ...... 68 3.2.2 Bioassay Design ...... 68 3.2.3 Data Analysis ...... 71 3.3 Results ...... 71 3.3.1 Aedes aegypti...... 71 3.3.2 Aedes albopictus ...... 73 3.4 Discussion ...... 75 3.5 Disclosures ...... 78 3.6 Figures ...... 79

4 SUB-LETHAL EFFECTS OF VOLATILE PYRETHROIDS ON RESISTANT AND FIELD STRAINS OF AEDES AEGYPTI AFTER BRIEF EXPOSURE DURATIONS ...... 84

4.1 Introduction ...... 84 4.2 Materials and Methods ...... 87 4.2.1 Insect Rearing ...... 87 4.2.2 Bioassay Design ...... 88 4.2.3 Data Analysis ...... 89 4.3 Results ...... 90 4.4 Discussion ...... 92 4.5 Disclosures ...... 96 4.6 Figures ...... 97

5 OLFACTOMETRIC COMPARISON OF VOLATILE PYRETHROID BARRIER FORMULATIONS USING MOSQUITO BEHAVIORAL RESPONSES ...... 102

5.1 Introduction ...... 102 5.2 Materials and Methods ...... 103 5.2.1 Insect Rearing ...... 103 5.2.2 Modular Wind Tunnel ...... 104 5.2.3 Procedural Validation ...... 106 5.2.4 Behavioral Assay ...... 108 5.2.5 Data Analysis ...... 110 5.3 Results ...... 111 5.4 Discussion ...... 113 5.5 Disclosures ...... 115 5.6 Figures ...... 116

6 EVALUATION OF VAPOR-ACTIVE PYRETHROIDS AS AN OUTDOOR RESIDUAL TREATMENT (BARRIER) AGAINST MOSQUITOES ...... 120

6.1 Introduction ...... 120 6.2 Materials and Methods ...... 123 6.2.1 Insect Rearing ...... 123 6.2.2 Aging Bioassays ...... 123 6.2.3 Field Study Area ...... 125 6.2.4 Surveillance ...... 126

6.2.5 Treatment ...... 127 6.2.6 Leaf Excision Bioassays ...... 127 6.2.7 Data Analysis ...... 128 6.3 Results ...... 129 6.3.1 Residual Longevity ...... 129 6.3.2 Treatment Surveillance ...... 130 6.4 Discussion ...... 133 6.5 Disclosures ...... 136 6.6 Tables...... 137 6.7 Figures ...... 138

7 EFFICACY OF METOFLUTHRIN IN RESIDUAL INSECTICIDE BLENDS FOR MOSQUITO CONTROL ...... 144

7.1 Introduction ...... 144 7.2 Materials and Methods ...... 147 7.2.1 Insect Rearing ...... 147 7.2.2 Site Evaluation ...... 147 7.2.4 Data Analysis ...... 149 7.3 Results ...... 150 7.4 Discussion ...... 155 7.5 Disclosures ...... 158 7.6 Figures ...... 159

8 CONCLUSIONS ON THE FUTURE OF VOLATILE PYRETHROIDS FOR INTEGRATED VECTOR MANAGEMENT ...... 166

APPENDIX ...... 170

LIST OF REFERENCES ...... 177

BIOGRAPHICAL SKETCH ...... 194

LIST OF TABLES

Table page

2-1 Comparative LC50 values of five volatile pyrethroids, delivered as a vapor, to pyrethroid susceptible strains of Aedes albopictus and Ae. aegypti...... 58

2-2 Comparative LC50 values of five volatile pyrethroids, delivered as a vapor, to pyrethroid susceptible strains of Culex quinquefasciatus and Anopheles quadrimaculatus...... 59

2-3 Comparative LC90 values of five volatile pyrethroids, delivered as a vapor, to pyrethroid susceptible strains of Aedes albopictus and Ae. aegypti ...... 60

2-4 Comparative LC90 values of five volatile pyrethroids, delivered as a vapor, to pyrethroid susceptible strains of Culex quinquefasciatus and Anopheles quadrimaculatus ...... 61

6-1 Daily environmental conditions for aging metofluthrin treated substrates...... 137

LIST OF FIGURES

Figure page

1-1 Hypothetical exposure outcomes of mosquitoes exposed to volatile pyrethroids...... 44

2-1 Simplified vapor bioassay exposure chamber...... 62

2-2 Perfumery strip saturation method ...... 63

2-3 Toxicant Relationship Diagram ...... 64

3-1 Overall design of the experiment with time estimates showing the duration of steps...... 79

3-2 Post transfluthrin vapor exposure decrease in egg dispersion across containers ...... 80

3-3 Post transfluthrin vapor exposure reduction in fecundity ...... 81

3-4 Multi-plate figure showing Aedes aegypti (L.) eggs following exposure of the parent female mosquito to sub-lethal concentrations of transfluthrin ...... 82

3-5 Multi-plate figure showing dissected female Aedes albopictus (Skuse) reproductive tracts after exposure to sub-lethal concentrations of transfluthrin and upon conclusion of oviposition bioassays seven days post-exposure ...... 83

4-1 Reduction in the mean percent usage of oviposition sites by Aedes aegypti (L.) after sub-lethal exposure to metofluthrin vapors...... 97

4-2 Reduction in egg yield and expression of multiple egg phenotypes among three strains of Aedes aegypti (L.) after exposure to sub-lethal concentrations of metofluthrin vapors for 60 seconds ...... 98

4-3 Multi-plate figure showing sub-lethal effects expressed by Aedes aegypti (L.) 1952 Orlando strain, St. Augustine wild type, and pyrethroid resistant Puerto Rican strain following 60s exposure of the parent female mosquito to metofluthrin vapors prior to blood feeding ...... 99

4-4 Multi-plate figure showing egg phenotypes occurring as sub-lethal effects in Aedes aegypti (L.) St. Augustine wild type, Puerto Rican strain, and 1952 Orlando strain following 60s exposure of the parent female mosquito to metofluthrin vapors prior to blood feeding ...... 100

4-5 Kaplan-Meier plots displaying decreases in larval survivorship across the post- bioassay period for treatments and controls of 1952 Orlando, St. Augustine wild type, and pyrethroid resistant Puerto Rican strain Aedes aegypti (L.) ...... 101

5-1 Diagramatic breakdown of a modular wind tunnel, designed of glass and metal, intended for olfactometry using volatile pyrethroids against mosquitoes ...... 116

5-2 Modular wind tunnel design that prevents chemical memory build-up across bioassays ...... 117

5-3 Procedural validation with a modular wind tunnel ...... 118

5-4 Mean behavioral response of Aedes albopictus (Skuse) after olfactometry bioassays conducted with volatile pyrethroid containing products ...... 119

6-1 Environmental chambers for housing treated wood substrates inside incubators ...... 138

6-2 Field sites used for application of barrier treatments ...... 139

6-3 Leaf bioassays where six leaves were removed in 10m transects from the treated vegetation of every field site ...... 140

6-4 Aging bioassay data visualizing percent knockdown after 20min or mortality after 24hr in adult female Aedes albopictus (Skuse) ...... 141

6-5 Leaf bioassay data visualizing the change in percent knockdown after 20min or mortality after 24hr exposure in adult female Aedes albopictus (Skuse) ...... 142

6-6 Percent reduction in the relative abundance of Aedes albopictus (Skuse) in the field after vegetation was treated with a high label rate for either metofluthrin, Onslaught Fast Cap, or a duplex treatment of Onslaught + metofluthrin ...... 143

7-1 Percent reduction in the relative abundance of Aedes albopictus (Skuse) in the field after vegetation was treated with a high label rate of either OneGuard, Sector, Hyperion, Onslaught Fast Cap, metofluthrin, or a duplex treatment of Sector/Hyperion/Onslaught + metofluthrin ...... 160

7-2 Leaf bioassay graphs visualizing percent knockdown after 20min or mortality after 24hr in adult female Aedes albopictus (Skuse) ...... 162

7-3 Aedes albopictus (Skuse) eggs diagnosed with reduced viability after removal from field sites treated with residual insecticide ...... 164

7-4 Reduction of Aedes albopictus (Skuse) viable eggs from field sites treated with residual insecticide containing metofluthrin ...... 165

A-1 Spectrum recording for broad spectrum ultraviolet lights used to decontaminate the modular wind tunnel of spatial repellent deposits ...... 170

A-2 Percent reduction in the relative abundance of the general host-seeking mosquito population in the field after vegetation was treated with either a low or high rate of metofluthrin, Onslaught Fast Cap, or a duplex treatment of Onslaught + metofluthrin.. 171

A-3 Percent reduction in the relative abundance of Aedes albopictus (Skuse) in the field after vegetation was treated with a low label rate of either metofluthrin, Onslaught Fast Cap, or a duplex treatment of Onslaught + metofluthrin ...... 172

A-4 Percent reduction in the general mosquito population in the field after low rate and high rate applications of OneGuard, Sector, Hyperion, Onslaught Fast Cap, metofluthrin, or a duplex treatment of Sector/Hyperion/Onslaught + metofluthrin ...... 174

A-5 Percent reduction in the relative abundance of Aedes albopictus (Skuse) in the field after vegetation was treated with a low label rate of either OneGuard, Sector, Hyperion, Onslaught Fast Cap, metofluthrin, or a duplex treatment of Sector/Hyperion/Onslaught + metofluthrin ...... 176

LIST OF ABBREVIATIONS

ABC transporter ATP-Binding Cassette; transporter proteins contained within cell membranes that move biological agents, native or non-native, into or out of the cell ad libitum According to pleasure; used to indicate when no specific schedule is abided for a protocol, instead being administered as needed or to achieve a certain outcome

Ae. Mosquito genus Aedes

AMCD Anastasia Mosquito Control District of St. Johns County, Florida

An. Mosquito genus Anopheles

ATP Adenosine Triphosphate; product of cellular respiration that is used in biological actions requiring energy input

CDC Centers for Disease Control and Prevention, United States

CHIKV Chikungunya virus, alphavirus

DDT Dichlorodiphenyltrichloroethane, an organophosphate banned from use in the United States in 1972

DEET Diethyltoluamide, the most common active ingredient in insect repellent. Developed in 1944 by the United States Department of Agriculture

DENV Dengue virus, flavivirus

Dv0.1/0.5/0.9 Diameter Volume; statistic representing a decimal value, between 0-1, which relates the volume proportion of a spray cloud to the drop diameter at which the cloud is made of drops equal to or smaller than this (ex: 20 micron) drop, and whose cumulative volume equals the proportion of interest (0.1, 0.5, 0.9).

EEEV Eastern equine encephalitis virus, alphavirus

EPA United States Environmental Protection Agency

ESI-MS/MS Electrospray ionization mass spectrometry with tandem mass spectrometry, used to diagnostically identify molecules from compounds; technique where high voltage is used to aerosolize a substance that is then fragmented using multiple selection processes

FL Fiducial Limit; calculated for means or proportions when the underlying distribution is logistic (s-shaped) with cutoffs at 0 and 100%

IRS Indoor Residual Spray; the act of applying an insecticidal chemical with long lasting residual to the indoor wall space of domiciliary structures

JEV Japanese encephalitis virus, flavivirus

Kdr Knockdown resistance gene; a heritable phenotype that is both used as a marker of a distinct genotype and is known to impart tolerance or resistance to the acute effects of pyrethroids and organophosphates

LC10/30/50/90 Lethal Concentration at which 10%, 30%, 50%, or 90% of subjects die, respectively; expressed as mg (or equivalent) of compound / ml (or equivalent) of application area

LD10/50/90 Lethal Dose at which 10%, 50%, or 90% of subjects die, respectively; expressed as mg (or equivalent) of compound / kg (or equivalent) of subject’s mass

L:D Light : Dark; the number of hours assigned to light and dark filled environment, expressed as a ratio of 24 hours

MWT Modular Wind Tunnel: an experimental hybrid of wind tunnel and olfactometer designs made entirely of glass and structural aluminum

PPB Parts per billion; a measure of concentration in air or water equivalent to 1 μg/L (or 10-9)

PPM Parts per million; a measure of concentration in air or water equivalent to 1 mg/L (or 10-6)

Ps. Mosquito genus Psorophora

RH Relative Humidity; typically expressed as the median percentage value with a representative error (example: 85±5%)

RO Reverse Osmosis water; water filtered through a semi-permeable membrane in a pressurized environment to remove ions, molecules, and large particulates

SLEV St. Louis encephalitis virus, flavivirus s.s. Sensu Stricto: Literally “in the strictest sense.” Used when referencing the originally described species that exists in a complex of still unclarified sub-species or unresolved taxonomic conflict.

Suffix, -thrin Any of the synthetically derived chemicals that are pyrethroids

Type-II Pyrethroid Pyrethroid containing an α-cyano group on a phenoxybenzyl ring alcohol; sodium channel modulator that triggers contractions with writhing body motions (Choreoathetosis) and salivation

Type-I Pyrethroid Pyrethroid containing chrysanthemic acid without an α-cyano group on a phenoxybenzyl ring alcohol; sodium channel modulator that triggers tremors (Tsyndrome)

ULV Ultra-Low Volume; a technology that shears liquid using a vortex air flow and constricted nozzle to disperse low volumes of insecticidal treatments as fine mist particles

USDA-ARS- United States Department of Agriculture, Agricultural Research CMAVE Service, Center for Medical and Veterinary Entomology

VmeD Volume Median Diameter; descriptive statistic that relates a particular droplet size (ex: 20 micron) being produced to the volume proportion of the cloud composed of drops equal to or small than the droplet size of interest.

WEEV Western equine encephalitis virus, alphavirus

WHO World Health Organization

WNV West Nile virus, flavivirus

YFV Yellow fever virus, flavivirus

ZIKV Zika virus, flavivirus

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy

CHARACTERIZATION OF VOLATILE PYRETHROIDS FOR MOSQUITO MANAGEMENT

Christopher Stephen Bibbs

August 2019

Chair: Phillip E. Kaufman Major: Entomology and Nematology

Spatial repellents pre-date most forms of modern public health pest management tools.

Today, volatile pyrethroids bear the standard for research in spatial repellency against mosquitoes. Volatile pyrethroids seldom have been explored beyond pushing insects away from hosts. Using guiding literature, a dose-response screening was conducted using flumethrin, prallethrin, metofluthrin, transfluthrin, and meperfluthrin against four mosquito species.

Polyfluorinated pyrethroids, such as transfluthrin and metofluthrin, were significantly more toxic to mosquitoes than prallethrin and flumethrin, which are used in commercial products for vector management. To elaborate on the findings, transfluthrin vapors were administered at sub-lethal concentrations against Aedes aegypti and Ae. albopictus, resulting in reduced skip-oviposition and a multi-faceted 50-75% reduction in fecundity. When replicated with metofluthrin, the results were echoed regardless of resistant and field phenotypes of Ae. aegypti. Additional larval mortality in the F1 resulted in a 20-40% drop in overall fitness of mosquitoes.

Novel behavioral analysis with a modular wind tunnel revealed that the experimental metofluthrin residual treatment was more volatile and considerably more potent than a blend containing prallethrin. Compared to 75% attraction upwind to host cues against the prallethrin-

containing product, the metofluthrin treatment resulted in less than 10% attraction upwind alongside a concert of repellency, disorientation, knockdown, and death of Ae. albopictus. Aging bioassays confirmed formulation longevity of 2-3 weeks with 2 cm of average weekly rainfall. In the field, blends where metofluthrin was added to other pyrethroid-containing residual products displayed improved longevity up to 3-4 weeks in spite of 4 cm average weekly rainfall and 20-

40% greater Ae. albopictus egg and adult reduction in the field than with unblended products.

Replicating the study with a larger cast of products supported that esfenvalerate and sumithrin blended with metofluthrin performed best, resulting in an additional 30-40% reduction in Ae. albopictus egg viability observed from rearing field collected eggs.

Volatile pyrethroids have greater impacts than spatial repellency alone. The progressive findings throughout study support a wide range of beneficial, supportive attributes that nor-trans- chrysanthemate pyrethroids can offer area-wide mosquito management strategies. Future work should emphasize creative implementation to capitalize on the potential of previously dubbed

‘spatial repellents.’

CHAPTER 1 INTRODUCTION

1.1 Preface

The global burden of arthropod-borne diseases has been summarized by Hill et al. (2005) at approximately 1.5 million annual deaths worldwide. This burden translates into heightened pressure on humans to protect themselves from vector contact. Traditionally, we think of this as killing the vector, but contemporary vector control weaves in the goals of long-term management, which concerns itself with sustainability, avoidance of pesticide resistance selection, and multi-faceted control mechanisms. The ideology that advocates prevention and multi-faceted approaches to intervention (control) in order to achieve long-term results is collectively referred to as integrated pest management. When this concept is applied to disease vectors, it is termed integrated vector management. When applied to mosquitoes, it is integrated mosquito management. The last two are essentially one in the same.

For integrated mosquito management, capitalizing on community buy-in of this integrated approach can be as important as the direct intervention tools available. For example,

Bodner et al. (2016) highlights that area-wide operations in mosquito management are encountering a growing number of hurdles. These can include limited financial options or privatized land and other types of inaccessible properties limiting capacity for vector habitat treatment. To counteract this, the World Health Organization (WHO) has recognized the need for recruiting the general populace in vector management (WHO 2015). However, leaning on citizens to supplement vector control is not a new tactic. Vector control programs, reported by

Leontsini et al. (1993) used education to recruit the citizen-base into removing mosquito habitat and using proper personal protection to combat Aedes aegypti (L.), which is a globally significant vector of yellow fever (YFV), dengue (DENV), Chikungunya (CHIKV), Zika

(ZIKV), and other viruses (Derraik and Slaney 2015, Ngoagouni et al. 2015, Wilson and Chen

2015). Averett et al. (2005) reported that source reduction and personal protection were employed similarly against the mosquitoes that transmit West Nile virus (WNV). Recent emphasis on container-inhabiting mosquitoes has reinforced the usage of citizen recruitment methods against Aedes albopictus (Skuse), which is regarded as the most invasive mosquito species in the world (Benedict et al. 2007, Bartlett-Healy et al. 2011). However, recruitment of the public into vector management is not consigned only to cultural control methods like sanitation, water awareness, and pest-proofing.

The presence of ZIKV in the United States in 2016 has led to nation-wide campaigns educating risk groups on avoiding transmission hot spots and the importance of using personal protective measures. Education programs and occupational safety programs for people at risk for mosquito-borne illness were urged by the Centers for Disease Control and Prevention (CDC) to employ repellent products to supplement vector control (CDC 2016). Yet not all products available to consumers are traditional topical repellents, which have served as a personal protection standard in vector control. Some products use volatile pyrethroid compounds, which are a subset of synthetic pyrethroids that transition easily into a gaseous state, and delivered as spatial repellents rather than traditional adulticides. These are available to consumers in emanators, vaporizer mats, burnable coils, and other means that have been shown by Xue et al.

(2012a; 2012b) and Revay et al. (2013) to prevent mosquito biting. These tools are believed to create a vector-free area using a single product, which can protect multiple persons in one application (Cook et al. 2007). The large number of commercial products available serves as an indicator of public desire for these volatile pyrethroids. These products bear labels that suggest

protection from vectors. Such labeling pairs with educational programs encouraging individuals to be responsible for protecting themselves (i.e. use a repellent).

Repellency has been studied frequently in volatile pyrethroids, including metofluthrin, transfluthrin, and prallethrin (Argueta et al. 2004, Lee 2007, Abdel-Mohdy et al. 2008, Achee et al. 2012b). Specific investigation by Achee et al. (2012b) found volatile pyrethroids present in a gaseous state deterred entry of mosquitoes into treated spaces. Lucas et al. (2007) documented that volatile pyrethroid compounds elicited an escape response in mosquitoes, expelling them from the affected air space. However, further evaluation by Ritchie and Devine (2013) found a disorientation response present in tested mosquitoes. This disorientation suggested that mosquitoes may be exiting the area due to poor spatial awareness, rather than a truly expellant effect. This concept has been supported by demonstrating that regardless of proximity to the host, a disoriented mosquito is unable to contact the host, regardless of whether the mosquito escapes successfully (Buhagiar et al. 2017a). The value of repellency on the whole has been questioned in research by Moore et al. (2007), indicating that rather than pushing vectors away from all hosts, they simply push them to the nearest unprotected host. This potentially invalidates the benefit of repellency when used on a community or population scale.

Volatile pyrethroids should not be dismissed as a sub-par alternative to topical repellents.

These compounds produce a wide range of effects, depending on the exposure sustained by the mosquito. The known range of effects can be summarized as: spatial repellency, which causes directional movement away from a stimulus and prevents the vector from entering an area; disorientation, which causes non-directed movement away from a stimulus preventing the vector from landing, probing, or biting; contact irritancy or excito-repellency, which causes directed movement away from a tarsal or gustatory stimulus and prevents the vector from landing,

probing, or biting through ataxia; knockdown, which display as tremors and subsequent severe ataxia due to nerve excitation and prevents the vector from flying, walking, or biting successfully; mortality, both with acute onset following knockdown and as a delayed response after the vector may have successfully escaped continued exposure.

These varied outcomes occur in concert or succession, leading to a barrage of progressive disabilities that ultimately interrupt and prevent the successful blood-feeding of a vector. Lee

(2007) reported that transfluthrin and metofluthrin repellency yielded rapid mortality for both volatile pyrethroids. Furthermore, metofluthrin incited not only the disorientation, as previously reported by Ritchie and Devine (2013) and Buhagiar et al. 2017a, but also knockdown and eventual mortality in Ae. aegypti (Lee 2007, Ritchie and Devine 2013, Buhagiar et al. 2017a;

2017b), which prevented the mosquitoes from escaping continued exposure. Investigations by

Bibbs and Xue (2015) took the OFF! Clip-on ® product, a spatial repellent device expressing vapor-phase metofluthrin, and observed knockdown and mortality against Ae. aegypti in a product sold specifically for the repellent market. Prallethrin formulated as a repellent also has been found to elicit mortality as reported by Abdel-Mohdy et al. (2008). Therefore, repellency, or disorientation as the case may be, combined with knockdown and mortality as described in multiple studies eliminates the possibility described by Moore et al. (2007) of the vector moving to a less protected host. This provides a basis to consider volatile pyrethroids as a tool that generates more than one end point to achieve the goal of bite prevention. This vein of thought highlights the need to better understand volatile pyrethroids, or ‘spatial repellents,’ as well as to refine our concept of what this tool can offer as a bite preventative.

1.2 Literature Review

1.2.1 Repellency

The goal of a repellent is to prevent a blood-feeding arthropod from making successful contact with its host. Topical repellents are applied directly to the potential bite site, which also requires the vector to come within close proximity of the application site. Furthermore, if the application is uneven or otherwise applied incorrectly an exploitable gap in protection may be present. It remains that interfering with contact between a vector and a host is fundamentally required in order to reduce infection rates (WHO 2009). Volatile compounds have been studied to provide more repellent options.

The principle of a volatile repellent is alluded to in ancient Greek writings describing the cultural practices of various native peoples. For example, Egyptian fisherman used aromatic oils from fish and castor bean plants in crafting repellent torches and bed nets (Rawlinson 1996).

Anecdotal evidence describe that certain grasses were burned to repel biting insects using the smoke. However, many of these cultural practices are pre-scientific, causing the methodology described in such anecdotes to be difficult to verify. In turn, although such practices appear superficially to be the use of crude repellents, the mechanisms and utility of volatile compounds for repellency has garnered interest only in more modern studies.

Successful volatile compounds have been dubbed spatial repellents for their quality of creating an air space free of vectors, namely mosquitoes (Achee et al. 2012a). Spatial repellents have the potential to remediate many concerns associated with using repellents. It is noted by

Sugiharto et al. (2016) that repellent use suffers a lack of uniformity in populations. Meaning, not everyone will apply them. This returns to the argument made by Moore et al. (2007) that hosts lacking a repellent will suffer the biting pressure instead. However, spatial repellents create a more uniform usage pattern due in part that only one individual needs to use the tool to protect

multiple hosts in the vicinity (Cook et al. 2007, Paz-Soldan et al. 2011). This also circumvents failures to apply the repellent to all potential bite sites or to reapply when needed, both of which are experienced with topical repellents. This incentivizes the development of volatile compounds that may offer spatial repellent benefits.

It is described by Cook et al. (2007) that a prerequisite of spatial repellents is the disruption of successful blood-feeding over a large range. This includes interfering with a vector successfully detecting, locating, and approaching the host. These principles can be employed without our complete understanding of the process. For example, DDT was shown with Ae. aegypti to elicit a directed movement away from the source without the mosquito needing to physically contact the chemical deposit (Greico et al. 2007) and hence the mosquito did not die from the insecticide encounter. This falls within the developing modern concept that spatial repellents stimulate directed movement away from the source without physical contact being required. However, a contact irritant would provoke directed movement away from the source after physical contact. Conversely, a toxicant chemical would impair or kill the target after physical contact (Cook et al. 2007, Greico et al. 2007, Achee et al. 2009). These definitions have caused compounds that may have been dismissed as toxicants to be re-evaluated (i.e. contact irritants and spatial repellents). Pyrethroids, for example, are synthetic organic compounds based on pyrethrum discovered in Chrysanthemum spp. plants (Barnes and Verschoyle 1974). Using the ultrastructure of pyrethrum, many pyrethroid compounds with much higher activity than pyrethrum are used as toxicants in insecticides. Pyrethroids now account for as much as 17% of global insecticide sales, making them the second most utilized chemical class available in today’s market (Sparks 2013), as well as a class containing several potential spatial repellents

(Argueta et al. 2004, Lee 2007, Abdel-Mohdy et al. 2008, Achee et al. 2012b).

The potential for certain pyrethroids to act as spatial repellents has been quantified for metofluthrin by Lucas et al. (2007) as causing 85% of Aedes canadensis (Theobald), 89% of Ae. aegypti, and 95% of Aedes vexans (Meigen) to actively leave the treatment site. Furthermore,

Achee et al. (2012b) showed metofluthrin deterred 58% of exposed Ae. aegypti from entering a test room. Transfluthrin forced up to 93% of wild Culex pipiens quinquefasciatus Say and

Anopheles gambiae Giles s.s. away from test subjects and into nearby sentinel traps as reported by Pates et al. (2002). This was corroborated by Ogoma et al. (2012), who reported 98% of

Anopheles gambiae arabiensis Giles escaping transfluthrin impregnated hessian strips by moving to a different portion of the test area and away from human subjects. This was replicated again using transfluthrin on long lasting variants of hessian strips, and over 90% repellency of

Mansonia spp., Culex spp., and An. arabiensis was recorded for 6-months (Ogoma et al. 2017).

Similarly, prallethrin as investigated by Abdel-Mohdy et al. (2008) yielded 100% repellency of a mixed group of mosquitoes in lab testing. Liu et al. (2009) exposed Culex tritaeniorhynchus

Giles and Cx. quinquefasciatus to prallethrin and observed 80.34% repellency. Each of these volatile pyrethroids are currently used in liquid and heated emanators available on the global market to consumers (Argueta et al. 2004, Lee 2007, Abdel-Mohdy et al. 2008, Liu et al. 2009,

Achee et al. 2012a, Xue 2012a; 2012b, Revay et al. 2013).

The evidence for repellency derived from volatile pyrethroids is not without contradiction. In a movement study, Rapley et al. (2009) documented that Ae. aegypti, under the influence of metofluthrin, still entered and left the room equally. This is contrary to the idea a spatial repellent should initiate directed movement away from the source of exposure. A more recent study by Ritchie and Devine (2013) did not observe Ae. aegypti to have any increase in escape response, effectively ignoring the repellent effect. This was found alongside an 87%

reduction in biting noted by Rapley et al. (2009) and 100% reduction in biting noted by Ritchie and Devine (2013). This was partly attributed to a disorientation effect, which resulted in what could be interpreted as repellency when only seen as the mosquito leaving the area. However, the findings by Rapley et al. (2009) have provided an example where disorientation can lead a mosquito to possibly ignore or not recognize the stimulus as well, which may confound interpretation of repellency studies. Disorientation was documented by Kawada et al. (2006) with

Ae. aegypti wherein the disorientation was classified as a positive attribute, interfering with the mosquito’s ability to find a host effectively. Interference with host seeking agrees with the studies in question that despite recording a variable repellent effect, a significant reduction in biting was observed. Investigation by Msangi et al. (2010) found a significant difference in repellency between different vector species. In their study, d-allethrin, a volatile pyrethroid, caused 92%-98% of Cx. quinquefasciatus to exit the study area, but only 60%-64% of An. gambiae exited the area under similar conditions. This same study also demonstrated a difference in reduction of actual blood feeding by the mosquitoes that did not exit the area, with Cx. quinquefasciatus being stopped 91% of the time as compared to 59% in An. gambiae.

Sathantriphop et al. (2014) expanded interspecies response comparison by testing permethrin, a Type-I pyrethroid, and deltamethrin, a Type-II pyrethroid, in non-contact assays against Ae. aegypti, Ae. albopictus, Anopheles minimus Theobald, and Cx. quinquefasciatus. In these assays, Ae. aegypti and Cx. quinquefasciatus were found to not be repelled by permethrin and deltamethrin. Anopheles minimus was not repelled by deltamethrin, but was successfully repelled by permethrin. Aedes albopictus expressed the highest sensitivity, being consistently repelled by both permethrin and deltamethrin and an accompaniment of other test compounds.

Interestingly, Ae. aegypti displayed low responsiveness across the non-pyrethroid compounds as

well, which included DEET, picaridin, citronella, and several botanical oils. The other mosquito species tested were repelled by the non-pyrethroid compounds. Culex quinquefasciatus low responsiveness, in contrast with studies such as in Msangi et al. (2010) with 92%-98% response, could be an expression of resistance, as it responded to the non-pyrethroid compounds. Aedes aegypti could be an example of resistance or species-based insensitivities. Differential responses due to either reason hampers the benefits of acute symptoms such as knockdown and mortality.

Findings by both Stanczyk et al. (2013) and Sugiharto et al. (2016) demonstrated that in repellency assays with DEET, insensitivities can develop in mosquitoes following repeated contact with a repellent chemical. This was found in mosquitoes as little as three hours after initial exposure to DEET (Stanczyk et al. 2013). In this case, the insensitivities were attributed as a learned state, citing examples from Kelly and Thompson (2000) and Mwandawiro et al. (2000) where mosquitoes would respond differently to olfactory cues in order to maximize feeding success. If spatial repellents such as metofluthrin and d-allethrin have variable response in targets, this could be behaviorally mediated as well. Contemporary research by Wagman et al.

(2015a) found similar insensitivities when testing transfluthrin on Ae. aegypti. In these assays, the behavioral plasticity of Ae. aegypti was deliberately measured following repeated exposures to transfluthrin volatiles and it was found that mosquitoes exposed to the chemical in consecutive days were less likely to be repelled. Wagman et al. (2015a) also linked a predisposition for developing insensitivity to a heritable phenotype known as kdr.

Behaviorally overcoming a stimulus may account for some differences observed in the studies by Rapley et al. (2009) and Msangi et al. (2010); however a key difference remains in the findings of insensitivity to DEET, a topical repellent, versus the insensitivities to transfluthrin, a spatial repellent. DEET is attributed as interfacing with mosquito targets through odorant binding

receptors (Stanczyk et al. 2013, Sugiharto et al. 2016). The disorientation effect noted by

Kawada et al. (2006) and Rapley et al. (2009) demonstrates the different mode of action for spatial repellents. Ritchie and Devine (2013), tested metofluthrin against Ae. aegypti and found disorientation, acute paralysis, and significant mortality. The sum effect of these contributed to their recommendation that spatial repellents be used against vectors. Rapley et al. (2009) also noted that mosquitoes may have entered a treated room despite active chemical, but that knockdown afflicted over 80% of the test subjects. Thus, although insensitivities may develop such as in Wagman et al. (2015a), the mosquito still runs high risks of neurological penalties such as disorientation, knockdown, and mortality (Buhagiar et al. 2017a). Therefore, understanding acute symptoms, even toxicity, now becomes essential for determining appropriate development of spatial repellent compounds.

1.2.2 Acute Symptoms and Toxicity

It should be revisited that the spatial repellents being discussed are volatile pyrethroids.

Trans-chrysanthemate pyrethroids, which include many volatile pyrethroids, degrade quickly in the environment as well as present low mammalian toxicity in compounds we know to be potent against insects (Elliot 1973, Miyamoto 1976). These factors have contributed to the aforementioned widespread availability of the chemical class. However, their mode of action is fundamentally different from compounds like DEET which act on odorant binding receptors.

Pyrethroids are sodium channel modulators, a neuro-toxicant which interferes with the uptake of sodium in the neuron and results in persistent depolarization of the cell membrane (Barnes and

Verschoyle 1974, Miyamoto 1976). In fact, the depolarization of the neuron results in observed effects like hyperexcitation. The insect consequently develops ataxia, which renders it flightless and moribund. The visible effect of the insect being rendered flightless is termed knockdown or

KD (Harrison 1951). It could be attributed that the behavioral avoidance of volatile pyrethroids

is simply avoiding harmful stimulus. Studies regarding metofluthrin, prallethrin, transfluthrin, and d-allethrin have elaborated on these other benefits of the volatile pyrethroids available to consumers.

Lee (2007) followed up their repellency assays with analysis to determine paralytic effects. Transfluthrin yielded up to 87% knockdown within 10-minutes of exposure to Aedes togoi (Theobald) and Ae. albopictus. Metofluthrin achieved up to 59% knockdown on the same species, within the 10-minute period. Other findings examine products already available to consumers, rather than single ingredients. For example, ThermaCELL devices containing d- allethrin caused 80%-100% knockdown within 30-minutes in Aedes taeniorhynchus

(Wiedemann), Aedes atlanticus Dyar and Knab, Psorophora columbiae Dyar and Knab, and

Psorophora ferox Humboldt (Bibbs et al. 2015). Follow-up studies using the OFF! Clip-on, which contains metofluthrin, also caused over 90% knockdown in Ae. aegypti (Bibbs et al.

2015). Support for the idea that volatile pyrethroids cause knockdown as an acute symptom is relevant to findings in previously discussed work. Returning to Rapley et al. (2009) and Ritchie and Devine (2013), Ae. aegypti was reported to enter a test room and, upon contact with metofluthrin volatiles, suffer from knockdown. This particular acute effect prevents not only the vector from contacting a host, but also prevents the vector’s escape from the area. This leads the vector to experience sustained exposure to the volatiles. Rapley et al. (2009) quantified this effect and found that 98% of the Ae. aegypti that failed to escape the treatment area subsequently died from exposure. This allows the knockdown effect to augment the mortality outcomes, given that mosquitoes will accumulate greater doses of chemical the longer they fail to leave the area of volatilization. Here is where it is significant to evaluate tools on combined attributes, rather than one outcome. Previously, these compounds were evaluated based on repellency. For those

vectors that are not repelled, the higher dose acquired by failing to avoid the area has a high probability of immobilizing the target. This stops the vector mid-flight, preventing bite contact.

More importantly, the continuous inability of the vector to escape the area increases the likelihood of death, thereby eliminating any further pathogen transmission potential of that individual. Thus, the mortality in and of itself becomes an endpoint. How significant, then, is the mortality that can be expected from using these volatile pyrethroids?

The study by Bibbs et al. (2015) using d-allethrin removed the mosquitoes from the exposure area after the 30-minute testing window. After removal from the chemical source, mosquitoes were held in the laboratory for 24h post exposure where 94%-99% of all tested mosquito species were observed to have died. In a similar study by Bibbs and Xue (2015) metofluthrin-exposed mosquitoes were removed from chemical exposure at time points ranging from 5-minutes to 60-minutes. Here, nearly 30% mortality was observed in mosquitoes removed after 15-minutes of exposure. Following 30- and 60-minute exposures mortality jumped to 97% and 100%, respectively. Prior work by Xue et al. (2012a) had determined dimefluthrin, meperfluthrin, and rich-d-transallethrin to be insecticidal when applied as a spatial repellent via mosquito coils. Anopheles albimanus Wiedemann and Ae. albopictus both expressed from 70%-

97% mortality across these different chemicals. Culex quinquefasciatus was hardier, as dimefluthrin and rich-d-transallethrin caused 60%-70% mortality despite meperfluthrin causing

92% mortality. Prallethrin is commercially available in mosquito coil products. In an evaluation by Katsuda et al. (2009), prallethrin was evaluated against Ae. aegypti individually expressing one of eleven genotypes, with some genotypes known to have reduced susceptibility to d- allethrin. All individuals in the eleven genotypes evaluated in a 30-minute exposure window died following exposure. The time until 100% mortality ranged from 11-minutes to 120-minutes.

These studies collectively support a wide range of volatile pyrethroid compounds generating significant mortality in vectors within a relatively short 30-minute period. Although mosquitoes successfully repelled by these compounds will not necessarily perish due to leaving the treatment cloud, individuals that succumb to knockdown will more reasonably be exposed to the lethal 30-minute window. However, Katsuda et al. (2009) approaches an important consideration with their work on multiple genotypes of the same species.

Not all mosquitoes will be repelled or knocked down by these vapor-active chemicals.

Katsuda et al. (2009) deliberately sampled a large area and acquired Ae. aegypti strains with lower susceptibility to d-allethrin, a spatial repellent with considerable repellency, knockdown, and mortality data reported (Msangi et al. 2010, Dame et al. 2014, Revay et al. 2013, Bibbs et al.

2015). The mechanism for this resistance has been documented by Wagman et al (2015a).

Resistant phenotypes develop when selective pressures are high; as pointed out with transfluthrin, a set of alleles dubbed kdr has been linked with a target’s ability to resist or tolerate the acute symptoms of pyrethroid exposure (Harrison 1951), and even repellency (Wagman et al.

2015a).

As discussed earlier, there are multiple spatial repellent products available on the market for consumers. If increased usage results from either successful marketing or because consumers prefer them to topical repellents, then resistance development is a realistic consequence to consider. Resistance development is compounded by pre-existing differences in sensitivities across species. Msangi et al. (2010) demonstrated a 30%-40% lower response in An. gambiae in both repellency and reduction of blood feeding as compared to Cx. quinquefasciatus when exposed to d-allethrin. Similarly, Xue et al (2012a) reported Cx. quinquefasciatus exhibited

20%-30% less mortality than did An. albimanus and Ae. albopictus when exposed to dimefluthrin or rich-d-transallethrin.

1.2.3 Sub-Lethal Effects

Sub-lethal effects are an additional layer of action against target vectors, and for the purposes of this review will include different effects of exposure that do not result in death. For example, if mosquitoes avoid acute effects, such as through successful avoidance of the active ingredient or through mechanisms like knockdown resistance or variations in sensitivity, these will be weighed alongside effects that are harmful, such as neurological impacts or behavioral outcomes that ultimately harm the mosquito. In understanding those sub-lethal effects, difficulties posed by variable sensitivity or resistance can be overcome. One can examine the patterns observed following the use of DDT, one of the flagship examples of enhanced selection pressure resulting in increased insect resistance expression (Harrison 1951). It has been argued that despite the extreme selection pressure on insects with the advent of insecticides, there are anomalous examples where resistance did not develop (Chareonviriyaphap 2012). For example,

Trapido (1954) reported that An. albimanus in Panama had been controlled using DDT for several years, and yet were equivalently as susceptible as a colony with no prior DDT exposure.

In teasing apart the cause, it was concluded that when mosquitoes were allowed adequate lapse of generations between DDT selection events, the population remained susceptible (Trapido

1954, Chareonviriyaphap 2012). In another example, Anopheles darlingi Root was shown to have not developed insecticide resistance in Brazil when surveyed in 1984 despite an active multi-year malaria control program that relied on DDT applications (Roberts et al. 1984).

Another study reported a lack of DDT resistance in western Thailand’s An. minimus population even when routine indoor residual spray (IRS) application of DDT generated resistance in other regions of Thailand (Chareonviriyaphap et al. 1999, 2001). Behavioral avoidance of direct

contact with DDT has been considered causative of this phenomena (Chareonviriyaphap 2012).

This is somewhat analogous to Wagman et al. (2015a) where an alternating test sequence of 24h and 48h rest periods for Ae. aegypti pre-exposed to transfluthrin found that Ae. aegypti resumed typical behaviors and susceptibilities when allowed 48h without exposure. Thanispong et al.

(2009) utilized two DDT-resistant field strains of Ae. aegypti to determine that physiological resistance negatively correlated with sensitivity to either contact irritancy or non-contact repellency. Yet, the Ae. aegypti population retained a sub-lethal behavior to avoid contact with the insecticide. This echoes the patterns from Brazil and Thailand in that behavioral modification is a separate event from target site or metabolic resistance. This same sub-lethal avoidance behavior would correspondingly relax the selection pressure on resistance and allow susceptible phenotypes to re-emerge (Roberts et al. 2000).

As with the above examples involving DDT, just because a physiological mechanism of repellency fails does not mean the behavioral aspect is unimportant. Outside of resistant phenotypes, though, a confounding factor in repellency has already been observed: disorientation. This disorientation is another form of sub-lethal effect. Recall that Wagman et al.

(2015a) determined that a group of Ae. aegypti displayed insensitivity to the repellent action of transfluthrin. This insensitivity also was linked to the kdr phenotype that imparts resistance to mortality (Harrison 1951, Wagman et al. 2015a). However, their method of evaluation used a high-throughput screening system that did not involve host contact as a way of measuring overall bite prevention. It is not conclusive in this study that the lack of repellent sensitivity necessarily translates to a failure of the compound.

Compare this to Rapley et al. (2009) which introduced a volunteer host into the experimental design. Rapley et al. (2009) reported that mosquitoes were as likely to enter treated

rooms as exit them while a host was present. This was attributed to disorientation with support for this interpretation being the lack of affected mosquitoes attempting to find the host in either treated or untreated rooms. Instead, they would retreat to harborage. If that harborage was in a treated area, knockdown and mortality would occur as a result of lengthy exposure times. If they left the treated area entirely, Rapley et al. (2009) postulated the escaped mosquitoes abandoned the treated area despite the presence of a host. Ritchie and Devine (2013) also noted this when conducting their own treated room studies. They added that the disorientation appeared to reduce flight speed, and again prioritized seeking harborage over host contact. Their reports included data that upon introduction of metofluthrin into the room, successful contacts with a host, even a brief landing, were negligible within the 10-minute evaluation time. Those mosquitoes that found harborage outside the treatment area were noted to readily escape into traps mounted on the building windows. They terminated treatment by forcefully evacuating the metofluthrin vapors for a 2h period after which it was noted mosquitoes resumed landing on hosts. This suggests that this disorientation was a physiological state and not a learned state.

Interrupting blood feeding behavior and ultimately success is another sub-lethal effect of chemical exposure that could factor into bite prevention. Hao et al. (2008) made the point that some compounds can alter the blood-feeding behavior of a vector. Their work with Ae. albopictus indicated some vapors from botanical compounds would impair the mosquitoes’ ability to find a host. It also was found that mosquitoes would have increased time until probing the host, attributed to delays in orientation and activation (Hao et al. 2008). Sugiharto et al.

(2016) exposed Ae. aegypti to DEET and observed reduced blood engorgement for the following

24h. Pyrethroids also have been shown to depress blood engorgement. Liu et al. (1986) reported that when d-phenothrin, d-allethrin, and tetramethrin were applied as spray mist droplets Ae.

aegypti had 50%-60% reduction in blood meal uptake. Liu and Georghiou (1987) found trans- permethrin significantly reduced blood engorgement with topical applications to Cx. quinquefasciatus. Adanan et al. (2005) used vaporizer mats to administer d-allethrin and prallethrin to mosquitoes. This study reported both chemicals depressed blood engorgement of

Cx. quinquefasciatus by 30% and Ae. aegypti by 70%. Ogoma et al. (2014a) evaluated volatile pyrethroids for deterrence, knockdown, mortality, and other sub-lethal effects. They concluded that blood feeding inhibition was the greatest observed effect, with An. arabiensis and An. gambiae suffering from 98% depression of engorgement by transfluthrin and 93% depression by metofluthrin exposure. Ultimately, reduced contact time upon attempted blood feeding may reduce blood-borne parasite transmission.

Delaying the onset of a blood-feeding response can enhance other beneficial effects, particularly mortality, that will lower pathogen transmission risk. Ogoma et al. (2014b) reported that transfluthrin did not prevent An. gambiae from being attracted to the host, but greater than

75% of mosquitoes that passed through the transfluthrin treatment had reduced blood feeding attempts for at least 12h. In their design, mosquitoes were removed from the testing area and offered blood meals in a lab setting. In more realistic applications, the delay they demonstrated provides more than enough time for sub-lethal effects to manifest or for the host to escape the area.

Interrupting blood feeding does translate to another benefit: reduction of oviposition success. Ogoma et al. (2014a) continued monitoring An. gambiae after depression of blood feeding and allowed successfully blood-fed, treated mosquitoes the opportunity to oviposit. They documented a 97% reduction in egg production related to transfluthrin and a 91% decrease related to metofluthrin exposure, as compared to unexposed mosquito treatments. This has been

contested in an indoor study in Australia in which a stationary emanatory containing metofluthrin did not reduce the reproductive fitness of either male or female Ae. aegypti

(Buhagiar et al. 2017b). This too could stem from differential sensitivities among species.

Earlier work used a more direct approach to look at oviposition deterrence. A wide array of experimental and commercial skin repellent compounds have been shown to deter Ae. albopictus from ovipositing in containers fitted with a repellent treated cloth (Bar-Zeev and Ben-

Tamar 1968) or repellent contaminated water (Xue et al. 2001; 2003; 2006). Furthermore, Xue et al. (2004) found that gravid mosquitoes denied the opportunity to oviposit through repellent treated water had reduced fecundity and increased hatchling mortality. Choi et al. (2016) examined oviposition effects from a different perspective. They exposed Ae. aegypti to transfluthrin vapors and found that mosquitoes were 10% more likely to visit ovicups as compared to unexposed mosquitoes. Additionally, treated mosquitoes were attracted to bacteria- baited oviposition containers twice as often, as compared to unexposed mosquitoes. They hypothesized that transfluthrin exposure increased grooming. This may have correspondingly changed the olfactory acuity of mosquitoes. This suggests that spatial repellents may interact with olfactory sensors and alter a mosquitoes’ ability to perform appropriately, if not necessarily through a mechanism of repellency as with DEET binding to odorant receptors.

Olfactory outcomes are still poorly studied for a wide range of chemicals and taxa, but it is recognized as a sub-lethal outcome of exposures to carbamates and organophosphates in agricultural pests (Dewer et al. 2016). For mosquitoes, prior investigations by Cohnstaedt and

Allan (2011) showed that permethrin and deltamethrin both impaired host seeking ability in Cx. quinquefasciatus, An. albimanus, and Ae. aegypti. They noted erratic flight patterns, decreased flight speeds, increased changing of direction, increased angle of turns during flight, and slower

initiation and termination of flights during exposure to host volatiles. This again supports the idea of disorientation in mosquitoes exposed to pyrethroid volatiles. In this case, the results were considered indicative of a compromise in the olfactory acuity of test subjects. Choi et al. (2016) suggested that an increased olfactory acuity may have resulted from excessive grooming to compensate for presence of chemical. The critical difference in Cohnstaedt and Allan (2011) was that their recording environment was restrictive, and may have resulted in differences in behavior that otherwise would have been observable in more open flight areas.

Unfortunately, interrupting or depressing a blood meal may not always result in a change in vectorial capacity. In the case of DEET, the observed reduction in blood engorgement was not accompanied by reduced landing and probing of hosts (Sugiharto et al. 2016). For certain pathogen transmission cycles this may be a detrimental combination of qualities, as there is evidence that mosquitoes that take sub-optimal blood meals are both still responsive to host cues

(Klowden and Lea 1978) and more likely to engage in multiple bite feeding patterns (Edman et al. 1975), both of which are likely to inflate vectorial capacity. It has been suggested that volatile pyrethroids, like metofluthrin, reduce landing rates through disorientation, but it is not clear at this time if that is a sufficient counter-balance to the vector’s re-feeding possibility.

1.2.4 Evaluation Methods

The effectiveness of volatile pyrethroids has been assessed through a few common designs. These variations are critical in the history of their development into spatial repellents.

The earliest methods replicate human structures, often dubbed a hut design. The principle involves having a point of attraction, such as a host or harborages, in an area of defined boundaries in which mosquitoes are allowed flight. The environment is freely navigable by the mosquitoes, but involves high rates of exposure to a treatment present inside the boundaries.

These could take place within literal structures (Pates et al. 2002, Lee 2007, Katsuda et al. 2009,

Rapley et al. 2009, Achee et al. 2012, Ritchie and Devine 2013, Ogoma et al. 2014a, Wagman et al. 2015a) in which the space is partitioned into multiple rooms. Variations are seen where defined borders still exist, but are not a dwelling or analogue of living space (Cohnstaedt et al.

2011, Ogoma et al. 2012, Ogoma et al. 2014b) with examples being tunnels or screened enclosures. All variations are consistent in that a known number of mosquitoes are within the defined boundaries alongside a measure of the success of insects in reaching the point of attraction, often a host. Behavioral effects are not severely limited, so factors such as avoidance of the treatment may be observed. This makes an informative design for objectives concerned with spatial components, like repellency, but poor where spatial components interrupt data collection, such as toxicity.

An alternative evaluation method is high-throughput screening. This is essentially a containerized testing environment in which the mosquito passes through or is contained within a device in order to expose the target to specific conditions. This began with containers composed of simple materials, such as paper, cloth, metal, and glass. where the insect is contained for maximum exposure (Roberts et al. 1984, Liu et al. 1986, Adanan et al. 2005, Abdel-Mohdy

2008, Stanczyk et al. 2013, Sathantriphop et al. 2014). Contemporary work has birthed a true high-throughput design with a rather specific construction, and allows optimal control of exposure time, dosage, and the allowance of contact or non-contact variants (Achee et al. 2009,

Thanispong et al. 2009, Wagman et al. 2015b). Regardless of type, these are defined by the insect’s inability to avoid treatment, and often involves close observation of the signs of exposure, making them more informative for sub-lethal effects, but less informative when requiring environments with competing stimuli. The majority of the discussed effects from repellency, acute symptoms, and sub-lethal effects have been generated by hut studies and high-

throughput studies. It can be gathered from their designs that it is difficult to account for competing stimuli and resilience to confounding spatial components.

Semi-field and field methods are employed as a response to the shortfalls of the prior two methods. These types of test allow competing stimuli and/or multiple stressors, such as environmental constraints, to factor into treatment outcomes. Semi-field methods are more controlled than field methods, and still allow for a known test group, approximated dosage, and are not impaired by the spatial components of testing (Bibbs et al. 2015, Bibbs and Xue 2015,

Obermayr et al. 2015, Buhagiar et al. 2017a; 2017b). Field studies are the most realistic method, but the most difficult to define and perform. Studies often take place where a treatment is employed against known pressures, such as bite contact, and it is then assessed for its ability to impede or remediate those pressures (Xue et al 2012a; 2012b, Revay et al. 2013, Dame et al.

2014). Treatments in these types of studies cannot be measured directly, as there is an uncontrolled input of targets. Instead, success is measured through proxies such as trap surveillance or recording successful contacts with a volunteer. Field designs are not limited in scope. They can incorporate multi-treatment evaluations and strategic implementation in informative ways.

The developmental strides made by the hut and high-throughput studies have allowed semi-field and field studies to show that spatial repellents can act as barriers, by deterrence and knockdown, to mosquitoes entering structures (Achee et al. 2012b, Menger 2015, Syafruddin et al. 2014, Wagman et al. 2015b). When used in tandem with traps or attractants, they can create push-pull systems by repelling mosquitoes into the traps or towards attractants, creating localized areas of mosquito absence (Revay et al. 2013, Syafruddin et al. 2014, Obermayr et al. 2015,

Wagman et al. 2015b). When combined with barriers or insecticides, such as with netting or

residual treatments, push-kill synergy is observed (Yuan and Huang 2014, Huho et al. 2015,

Paliga et al. 2015). These findings demonstrate that the known effects of spatial repellents can be put to use in integrated management strategies. New strategies, or enhancements to existing strategies, could be developed around other known benefits. For example, compounds extracted from various plants have deterred Ae. albopictus, Ae. aegypti, and Cx. quinquefasciatus from laying eggs at otherwise suitable oviposition sites (Swathi et al. 2010, Yu et al. 2015). Spatial repellents have been hinted at impairing mosquito oviposition (Ogoma et al. 2014a, Choi et al.

2016). If this were reinforced with further findings, a new strategy could develop where volatile pyrethroids are used to limit access to containerized breeding sites normally occupied by Ae. aegypti or Ae. albopictus.

A critical deficit in the evaluation of spatial repellents is in the epidemiological impacts following their implementation. One notable example to address this deficit is presented in a study using spatial repellents at each residence in a village suffering from active malaria transmission. In this study, the use of spatial repellent coils reduced malaria incidence by 52%

(Syafruddin et al. 2014). Despite this, targeted research to support whether spatial repellents can improve epidemiological outcomes is sparse. This is acknowledged in the current paradigms of the WHO that necessitates researching epidemiological impacts of spatial repellents for adoption of this tool into vector management protocols (Achee et al. 2012a). In these paradigms, it is described that spatial repellents have the benefit of not requiring lethality to be of utility.

Concurrently, the range of effects, lethal or otherwise, produced by spatial repellents is entirely the reason that the broader impacts of their mainstream use are brought into question (Achee et al. 2012a). Furthermore, despite the breadth of work presented here, studies almost never measure the airborne concentrations of active ingredient when documenting effects such as

repellency or disorientation. Only Achee et al. (2012b) reported a basic air quality measurement as a response to using a metofluthrin mosquito coil, having found outdoor/indoor concentrations of 0.028-0.088μg/0.009-0.029μg after 6h during hut studies. This makes it difficult to compare some vector outcomes when using spatial repellents, and adds complexity to predicting epidemiological relevance.

1.3 Objectives

The large body of work through decades of vector management research reflects that the experimental designs across these studies, as well as which chemicals, the form of the products, the vectors chosen, and even the genotype of the targets are all variable. This makes comparisons of the current studies to those in the literature difficult and not necessarily linear. The differences between the Rapley et al. (2009) and Wagman et al. (2015a) studies, as discussed in the sub- lethal effects section above, serve to illustrate this point. These differences are further reinforced in the discussion of examples including Cohnstaedt and Allan (2011) and Choi et al. (2016) with their opposed findings regarding olfaction impacts. Without the ability to directly compare, it remained challenging on how best to move forward. Regardless, spatial repellents have been repeatedly advocated to combat disease vectors (Xue et al 2012a; 2012b, Ritchie and Devine

2013, Dame et al. 2014, Bibbs et al. 2015, Bibbs and Xue 2015; 2016). This is in part because of abundant documentation supporting both simple mixtures and formulated products causing knockdown and mortality in mosquitoes. However, the efficacies presented do not qualify as a measure of lethal dose or lethal concentration. There is no concerted effort to deliberately evaluate these compounds for their toxicity because of the existing paradigm that volatile pyrethroids are repellents. Given the evidence at present, it may be better to step away from the idea of a spatial repellent and instead embrace that these are vapor-active insecticides. Aside from toxicity, many sources bring to light a suite of partially explored behavioral and sub-lethal

effects, or those occurring at the LD10-25/LC10-25 or lesser parts of a dose response curve. It is easy to imagine this string of potential outcomes along the dose response as a kind of spectrum that corresponds to different encounters between the mosquito and the toxicant. This concept is summarized in Fig 1-1, and is a hypothetical rendering of how the attributes discussed in this review may pertain to a flowchart of potential outcomes when a mosquito encounters a spatial repellent volatile pyrethroid. It is not clear at this time whether different volatile pyrethroids can be expected to generate the same effects. Regardless, these and similar attributes represent a potential that ultimately could change how these products are developed and deployed.

In summary, volatile pyrethroids, or spatial repellents, have multiple effects on mosquito vectors, but are currently only used one-dimensionally for spatial repellency. These tools occupy a unique niche of being spatial, yet both insecticidal and safe enough to use on your person.

There lies in this a great potential for incorporating volatile pyrethroids into vector abatement efforts, even if only by encouraging their use by the at-risk populace. Because of this, volatile pyrethroids need to have their toxicity, sub-lethal effects, and operational efficacy described to maximize their utility for preventing pathogen transmission.

1.4 Discolsures

This article was published in the Entomological Society of America Journal of Integrated

Pest Management in 2017, volume 8, page 21. Publication of this review was funded by the

University of Florida Open Access Publishing Fund.

1.5 Figures

Figure 1-1. Hypothetical exposure outcomes of mosquitoes exposed to volatile pyrethroids.

CHAPTER 2 VAPOR TOXICITY OF FIVE VOLATILE PYRETHROIDS AGAINST FOUR MOSQUITO VECTOR SPECIES

2.1 Introduction

Anthropophilic mosquitoes are the primary driver of the United States’ response to emerging tropical pathogens, such as Zika virus (ZIKV), which includes messaging for personal protective use (CDC 2016). The corresponding vectors of ZIKV, Aedes aegypti (L.) and Aedes albopictus (Skuse), are invasive, diurnal, container-inhabiting mosquitoes that are the respective primary and secondary vectors of dengue virus (DENV) and chikungunya virus (CHIKV)

(Derraik and Slaney 2015, Ngoagouni et al. 2015, Wilson and Chen 2015, Offerdahl et al. 2016).

While the malaria parasite is no longer endemic to North America, it is occasionally reintroduced by infected travelers (CDC 2003), with the principle vector being Anopheles quadrimaculatus

Say (Rutledge et al. 2005). Co-occurring with these other threats is one of the most widespread encephalitis virus vectors in the U. S., Culex pipiens quinquefasciatus Say (Day and Stark 1999).

This pollution-tolerant vector of WNV, SLEV and other encephalitis viruses develops in ditches, storm water runoff, and urban drain infrastructure throughout the Americas (Lumsden 1958,

Subra 1981, Day and Stark 1999, Noori et al. 2015).

As mosquitoes and pathogens are introduced into new locations, the increasing risk of emerging disease has led to an emphasis on the individual’s responsibility to protect themselves

(CDC 2016). Topical repellents are the cornerstone of personal protective measures against mosquitoes, but other tools compete for the attention of consumers. Spatial repellency is a relatively old strategem (Christophers 1947), with product development catching up in the modern age to find a variety of vaporizers, emanators, mats, and coils that are now available worldwide for use against mosquitoes (Argueta et al. 2004, Lee 2007, Abdel-Mohdy et al. 2008,

Liu et al. 2009, Achee et al. 2012a, Xue et al. 2012a; 2012b, Revay et al. 2013). The

distinguishing feature of these tools, compared to the myriad of other sprays and aerosol products, is that spatial repellents deliver active ingredients as a vapor rather than droplets that encapsulate a toxicant. This vapor diffuses across open space, and the cloud either repels or kills mosquitoes that are exposed (Xue et al 2012a; 2012b, Revay et al. 2013, Ritchie and Devine

2013, Dame et al. 2014).

Notable spatial repellent ingredients include vapor-active pyrethroids with polyfluorinated alcohols. Metofluthrin is one such pyrethroid, and developmental testing yielded over 90% mosquito mortality within an hour of exposure to 1.04 mg/m3 emanated from paper strips (Rapley et al. 2009). This method was further developed into latticed plastic diffusers

(Kawada et al. 2006). Metofluthrin is now commercially available in mosquito coils and portable blower fans, which have been reported to produce up to 95% mortality in Ae. aegypti (L.) in semi-field tests (Bibbs and Xue 2015). Transfluthrin is used outside the U. S. in several consumer tools, including mosquito coils as well, and predates metofluthrin in commercial development (Naumann 1998). When delivered through a blower fan at a rate of approximately 1 mg/m3, transfluthrin variable mortality in Ae. togoi (Theobald), Anopheles arabiensis (Giles), and Ae. aegypti (Lee 2007, Salazar et al. 2013, Andrés et al. 2015). 10 Meperfluthrin is a lesser known polyfluorinated pyrethroid, but was reported to caused 97%, 97%, and 92% mortality when delivered through a mosquito coil against An. albimanus Wiedemann, Ae. albopictus

(Skuse), and Culex quinquefasciatus Say, respectively (Xue et al. 2012a).

There are other pyrethroid active ingredients that could be candidates in spatial repellent products; but, these are currently used in different markets. Prallethrin is a non-fluorinated type I pyrethroid that is registered in commercial adulticides for mosquito management. It is formulated for thermal fogging or ultra-low volume cold aerosols, but also is dispersed as a

vapor from mosquito coils, wall plug-ins, and treated fabrics (Abdel-Mohdy et al. 2008). There can be crossover in effectiveness of spatial repellents against different vectors. For example, a recent study has shown vapors of metofluthrin and d-allethrin, which is a non-fluorinated pyrethroid, both repelled and killed ticks Conversely, flumethrin is a type II pyrethroid that is used primarily to repel or kill ticks on domestic animals (Fourie et al. 2015). This pyrethroid has not been tested against mosquitoes, but flumethrin has been demonstrated to repel and kill Sand flies in the genus Phlebotomus Loew (Jalilnavaz et al. 2016).

The mortality recorded in prior studies is consistent, but measurements were performed in environments with limited control of exposure (Achee et al. 2012b, Ogoma et al. 2012, Revay et al. 2013, Ogoma et al. 2014a) or in a way that did not exclude mosquitoes from physical contact with the compound (Adanan et al. 2005, Katsuda et al. 2009, Thanispong et al. 2009,

Sathantriphop et al. 2014). Mortality focused studies also have not attempted to characterize the range of outcomes on a dose response curve (Lee 2007, Rapley et al. 2009, Ritchie and Devine

2013, Wagman et al. 2015a, Choi et al. 2016). As a result, it is difficult to determine if spatial repellent compounds may be tenable as mosquito adulticides despite numerous reports of the active ingredients killing mosquitoes. Therefore, this study will screen prallethrin, transfluthrin, metofluthrin, meperfluthrin, and flumethrin against adult, female Ae. aegypti, Ae. albopictus, Cx. quinquefasciatus, and An. quadrimaculatus. By screening in one experimental design, lethal concentration (LC) response curves can be generated to identify which of these chemicals is widely active against key mosquito models.

2.2 Materials and Methods

2.2.1 Insect Strains

Pyrethroid susceptible mosquito strains were provided by the United States Department of Agriculture, Agricultural Research Service, Center for Medical, Agricultural, and Veterinary

Entomology (USDA-ARS-CMAVE) in Gainesville, Florida. These strains included 1952

Orlando, FL, strain Ae. aegypti; 1992 Gainesville, FL, strain Ae. albopictus; 1952 Orlando, FL, strain Cx. quinquefasciatus; and 1952 Orlando, FL, strain An. quadrimaculatus. Mosquito strains were not exposed to insecticides prior to evaluation and were not supplemented with wild type introductions to the colonies. Rearing conditions consisted of 26 ± 1°C, 85 ± 5% relative humidity (RH), with a photoperiod of 14:10 (L:D). Batches of 2,000 eggs were placed in larval pans containing 2,500 ml of reverse osmosis (RO) purified water. Larvae were fed 1–3 g of liver and yeast mixture at a 3:2 ratio ad libitum in a 50-ml suspension. Adult mosquitoes were kept in flight cages containing separate supplies of 10% sucrose solution and RO water. Across all species, subjects used in experiments were non-blood-fed, 5–7d old female mosquitoes.

2.2.2 Chemicals

Technical grade 98.2% prallethrin (32917 Pestanal, Sigma-Aldrich Co. LLC, St. Louis,

MO), 95.7% flumethrin (N-13139, Chem Service, Inc., West Chester, PA), 99.5% transfluthrin

(N-13626, Chem Service, Inc., West Chester, PA), 99.8% meperfluthrin (32065 Pestanal, Sigma-

Aldrich Co. LLC, St. Louis, MO), and metofluthrin were selected for this test. Technical grade metofluthrin was not available through a commercial supplier, therefore metofluthrin was extracted from OFF!® Clip-on over-the-counter refill packs (31.2% metofluthrin, S. C. Johnson

& Son, Racine, WI) using pentane. Extracts were fractionated using automated flash chromatography (CombiFlash Rd 200i, Teledyne ISCO, Lincoln, NE) with simultaneous electrospray ionization mass spectrometry (ESI-MS) (Expressions CMS, Advion, Inc., Ithaca,

NY, USA), using methanol:formic acid (99.9:0.1) as a mobile phase. Fractions were delivered using pentane as the non-polar solvent and ethyl ether as the polar solvent at a 10 ml/min flow rate and a 5 ml peak runtime. Solvent was reduced in a rotary evaporator and the resultant technical grade product was analyzed for purity using gas chromatography mass spectrometry.

Fractions rating at least 95% purity were used in subsequent steps. Each technical grade pyrethroid was serially diluted in acetone to create screening concentrations of 5.00%, 1.00%,

0.50%, 0.10%, 0.05%, and 0.01% solutions by weight, which were aliquoted into amber borosilicate vials (14-955-331, Thermo Fisher Scientific, Hampton, NH), wrapped in paraffin film, and stored at −80 for a maximum of 5d. Up to seven additional concentrations, different for each chemical, were selected with respect to the initial six range-finding dilutions, for a total of up to 13 concentrations that were replicated four times to collect sufficient data to determine

LC50 and LC90 values for each toxicant.

2.2.3 Fumigant Bioassay Design

Test cages (Fig 2-1.) were single-use 473 ml clear polypropylene snap-lid cups (MN16-

0100, Dart Container Corp, Mason, MI) with the lid modified to have a central 20-mm diameter opening. Twenty female mosquitoes of a single species were aspirated into each container. Filter paper strips (Grade 1 MFR# 28413934, Whatman PLC, Little Chalfont, UK) were cut into 5-mm widths and 40-mm lengths and pleated every 5 mm before being treated with 40-μl of a chemical solution (Fig 2-2). Treated strips were allowed 6-min drying periods before transfer into a mesh bag (Nylon Tulle No: 147356, Falk Industries, Inc., New York, NY) that was suspended within the test cage through the hole in the modified lid (Fig 2-1). The hole was then sealed to prevent vapor escape during testing. One concentration of a single chemical was used in each test cage, with a newly treated strip used for each replicate. Controls were strips treated with only acetone.

Test cages were stored in an incubator (Precision Mo: 818, Thermo Fisher Scientific, Hampton,

NH) to maintain 26 ± 1°C, 85 ± 5% RH, with a photoperiod of 14:10 (L:D) for the duration of data collection. Treatments were discarded from all cages after 2h and replaced with a cotton ball soaked with a 10% sucrose solution.

Residual activity data were collected by allowing mosquitoes exposed to vapors to remain in the original testing containers for the 24h test duration. Mosquitoes in a prone position and suffering from ataxia that prevented proper upright resting, walking, and flight were considered moribund, i.e., knockdown (KD). Mosquitoes in a prone position and rigidly immobilized were considered dead. The mortality scored in a test cage was comprised of the total combined score of moribund and dead mosquitoes. Mortality was scored at 2h, 4h, and 24h post exposure. Scores at 2h are considered indicative of KD, while 4h and 24h would allow some interpretation of metabolic recovery.

The persistence of vapors in the test container despite removal of the treated strips is less informative to the possibility that mosquitoes may escape continuous exposure to vapors in practical settings. Therefore, a series of recovery assays were performed following the above test cage construction and exposure procedures to assess recovery from vapor exposure. The holding conditions deviate in that mosquitoes were transferred to separate, untreated, single-use test cages after a 2h vapor exposure to emulate escape from the vapors. Mortality was scored by the same procedures. Mortality was recorded for these insects only after 24h. This was repeated in at least four replicates for all mosquito species and chemical concentrations to assess potential for metabolic recovery after treatment.

2.2.4 Data Analysis

Probit analyses were performed in PoloPlus (Version 1.0, LeOra Software LLC, Cape

Girardeau, MO) to derive descriptive statistics and dose responses of prallethrin, flumethrin, transfluthrin, metofluthrin, and meperfluthrin for Ae. aegypti, Ae. albopictus, Cx. quinquefasciatus, and An. quadrimaculatus at each repeated measure of time. A minimum of four replications with at least 2,080 individuals for each mosquito species were used per chemical to generate LC50 and LC90 values with a 95% fiducial limits (FL), expressed in m/v

(mass/volume or g/100 ml). Data were discarded if control mortality exceeded 10% within a replicate. Probit analysis included correction for control mortality using Abbott’s formula

(Abbott 1925) for all control mortality below 10%. If the lower FL and upper FL of two LC values did not overlap then the difference was considered significant. Variation with respect to chemicals and species was analyzed in JMP 13.1.0 (SAS Institute, Inc., Cary, NC) using a correlated sample analysis of variance (ANOVA). Vapor pressure was used as a contextual factor for toxicity by conducting correlation analysis in JMP.

2.3 Results

There was a positive correlation between vapor pressure and toxicity (r = 0.26, p <

0.0183) with a R2 of 0.069. When excluding flumethrin from the analysis, whose vapor pressure was lower by a factor of 1,000 from the other compounds, there was no correlation between vapor pressure and toxicity. The LC50 and slopes of the five pyrethroids are reported for Ae. albopictus and Ae. aegypti in Table 1. Data in Table 1 fall generally into three potency groups.

Transfluthrin and meperfluthrin had similar activity against Ae. albopictus and Ae. aegypti.

Metofluthrin and prallethrin are approximately 10-fold less active, and flumethrin is at least 20- fold less active than metofluthrin and prallethrin. Moreover, these relationships hold for Ae. albopictus across all time points. Activity was similar when comparing data between the 2h exposure group and the recovery group that was exposed for 2h followed by mosquito transferal to untreated containers until a 24h mortality reading. The activity against Ae. aegypti follows the same general pattern, but with more variability within the three potency groups. Responses to transfluthrin and meperfluthrin were comparable between Ae. aegypti and Ae. albopictus across all time points. There was a statistically significant effect of chemical on the mortality of Ae. albopictus (F(4, 231) = 15.86, p < 0.0001) and Ae. aegypti (F(4, 227) = 9.10, p < 0.0001). Non- overlap of upper and lower FL indicated that meperfluthrin was the most active against Ae.

aegypti within 2h, 4h, and recovery exposures, and against Ae. albopictus with the 2h exposure only. As indicated by the overlap of fiducial limits, meperfluthrin and transfluthrin were considered not significantly different from each other for Ae. aegypti 24h exposure and Ae. albopictus exposures at 4h, 24h, and recovery. Metofluthrin was significantly more active against Ae. aegypti at all exposures and Ae. albopictus at 2h. However, with Ae. albopictus metofluthrin was not significantly different from prallethrin at 4h, and prallethrin was significantly more active than metofluthrin at 24h and the recovery group. Mosquito mortality to each chemical differed significantly between species (F(3, 38) = 0.97, p < 0.0001). Notably, metofluthrin was more active against Ae. aegypti than Ae. albopictus at all exposures. Flumethrin was more active against Ae. aegypti at 4h and 24h exposures, while the other compounds showed variable toxicity between species depending on exposure duration. Slope values were generally

2–3 for Ae. albopictus, except for flumethrin and transfluthrin and at 24h, which approached 4 and 8, respectively (Table 1). Slope values for Ae. aegypti were less, typically 1–2, except for transfluthrin, which had uniformly the greatest slope values.

Table 2 shows the LC50 and slopes of all compounds for Cx. quinquefasciatus and An. quadrimaculatus. The same three potency groups hold for Cx. quinquefasciatus: transfluthrin and meperfluthrin; followed by metofluthrin and prallethrin with approximately 10-fold less activity; and flumethrin with at least 20-fold less activity than the former group. Although assigned to the same potency group, prallethrin was more active than metofluthrin against Cx. quinquefasciatus. As in Table 1, the data for the 2h exposure group and 2h exposure then recovery group were similar. For An. quadrimaculatus, the potency groups shifted in that transfluthrin, meperfluthrin, and metofluthrin were similarly the most active. Prallethrin was approximately 10-fold less active, and flumethrin was approximately 100-fold less active than

prallethrin. This trend was consistent across all exposures for An. quadrimaculatus. There was a statistically significant effect of chemical on the mortality of Cx. quinquefasciatus (F(4, 251) =

7.72, p < 0.0001) and An. quadrimaculatus (F(4, 247) = 29.41, p < 0.0001). When comparing upper and lower FL, transfluthrin and meperfluthrin were not significantly different at 2h and 24h exposures with either Cx. quinquefasciatus or An. quadrimaculatus. However, transfluthrin was most active at 4h and the recovery group against Cx. quinquefasciatus. Conversely, An. quadrimaculatus was significantly more sensitive to meperfluthrin at 4h and the recovery group.

Prallethrin was more active than metofluthrin at 2h, 24h, and recovery groups against only Cx. quinquefasciatus. Mosquito response to compound differed significantly between species (F(3, 38)

= 175.80, p < 0.0001). Transfluthrin and meperfluthrin were comparable in activity across both species and all time points. Prallethrin performed similarly against both Cx. quinquefasciatus and

An. quadrimaculatus, while metofluthrin was significantly more active against An. quadrimaculatus at all time points and flumethirn was much less active against An. quadrimaculatus at all time points. Slope values were generally 1–2 for Cx. quinquefasciatus and An. quadrimaculatus and there were no strong outliers (Table 2).

The LC90 of all compounds for Ae. albopictus and Ae. aegypti are reported in Table 3.

For both species at the 2h, 4h, and 24h recordings, transfluthrin and meperfluthrin were still the most active. Metofluthrin and prallethrin belonged to the second potency group, generally 10- to

30-fold less active, depending on exposure. The lowest potency was observed for flumethrin, which was several hundred fold less active against Ae. albopictus and over 1000-fold less active against Ae. aegypti. There was a statistically significant effect of chemical on the mortality of Ae. albopictus (F(4, 231) = 15.86, p < 0.0001) and Ae. aegypti (F(4, 227) = 9.10, p < 0.0001).

Meperfluthrin and transfluthrin were generally equivalent in activity across all time points

against both Ae. aegypti and Ae. albopictus. Metofluthrin and prallethrin also were equivalent, but significantly less active than meperfluthrin and transfluthrin, whereas flumethrin was much less toxic. Mosquito response to chemical differed significantly between species (F(3, 38) = 0.95, p

< 0.0001), but only consistently across all exposures with flumethrin, which was 100-1,000-fold more active against Ae. albopictus at 2h, 24h, and recovery groups versus Ae. aegypti.

Culex quinquefasciatus and An. quadrimaculatus LC90 values and fiducial limits are shown for all compounds in Table 4. The three potency groups indicated for Cx. quinquefasciatus in the LC50 data also apply to the LC90 data. Transfluthrin and meperfluthrin were most active. Metofluthrin and prallethrin form the next group, with generally 10-fold less activity. Flumethrin ranks last with over 1000-fold less activity. As before, these relationships hold for Cx. quinquefasciatus across all exposures. However, An. quadrimaculatus has different groupings: transfluthrin, meperfluthrin, and metofluthrin all had similarly high activity; prallethrin separates from these compounds with generally 10-fold less activity; and flumethrin is lowest with as much as 10,000-fold less activity. There was a statistically significant effect of chemical on the mortality of Cx. quinquefasciatus (F(4, 251) = 7.72, p < 0.0001) and An. quadrimaculatus (F(4, 247) = 29.41, p < 0.0001). Against Cx. quinquefasciatus, transfluthrin and meperfluthrin were equivalent across all exposures. The same is true for prallethrin and metofluthrin, but this group was significantly less active than transfluthrin and meperfluthrin.

Anopheles quadrimaculatus responded equally to transfluthrin and meperfluthrin across all time points. However, metofluthrin did not significantly differ from meperfluthrin at 4h and 24h, but transfluthrin was significantly more active than metofluthrin at the same exposures. Metofluthrin was equivalent to transfluthrin and meperfluthrin in the recovery group. Chemical activity differed significantly between species (F(3, 38) = 163.84, p < 0.0001). Flumethrin was more

uniformly active against Cx. quinquefasciatus than with An. quadrimaculatus, whereas the reverse was true for metofluthrin. All other responses were comparable between these two species.

2.4 Discussion

Transfluthrin, meperfluthrin, and metofluthrin (polyfluorinated type I pyrethroids) consistently killed mosquitoes at lower concentrations than flumethrin (monofluorinated type II pyrethroid) and prallethrin (non-fluorinated type I pyrethroid) (Fig. 2-3), the last of which is used for mosquito adulticiding. Poor performance was observed with flumethrin across all species, with the lowest observed mortality in Ae. albopictus. According to the United States

Environmental Protection Agency (EPA) chemical models, the vapor pressures of each compound are predicted as 1.32x10-5 mm Hg for transfluthrin, 2.94x10-6 mm Hg for meperfluthrin, 3.83x10-5 mm Hg for metofluthrin, 2.78x10-6 mm Hg for prallethrin, and 1.76x10-

9 mm Hg for flumethrin. It is concluded that the lower vaporization pressure with flumethrin in our study was the cause of the reduced efficacy.

The LC50 is less variable when comparing toxicant efficacies, and by this metric meperfluthrin generally performed as well or better than transfluthrin. However, the LC90 is a more relevant measure for future product development because the target approaches the required 95% efficacy to meet federal testing guidelines (EPA 2009). Using the LC90, meperfluthrin and transfluthrin were not significantly different in toxicity for the four species, reflecting the higher toxicity slope value for transfluthrin. Interestingly, transfluthrin and meperfluthrin consistently outperformed the other chemicals at the LC90 level against all mosquito species, except their toxicity was similar for metofluthrin against An. quardimaculatus.

This difference was especially clear in recovery assays, without a significant decrease in toxicity relative to 2h exposures. This comparison implies relatively equivalent performance with or

without persistent exposure to the treatment chamber. The equivalence could be attributed to the extent of halogenation on the alcohol moiety of transfluthrin and meperfluthrin (Fig. 2-3), which prevents oxidation by cytochromes P450 enzymes at these sites. Metofluthrin is registered in the

United States as a spatial repellent and shares this structural attribute on the alcohol, but was less active against all mosquito species except An. quadrimaculatus. However, additional factors, such as volatility, target site potency, and efficacy could also contribute to differential activity.

Cuticle penetration rates may be one factor that contributes to differential activity.

Penetration is not equivalent across different parts of a mosquito, even for topical application

(Aldridge et al. 2016, Aldridge 2017). Furthermore, penetration may be reduced through thickened epicuticle or an increased abundance of cuticular hydrocarbons, such as in a wild, pyrethroid-resistant population of Anopheles gambiae Giles (Balabanidou et al. 2016).

Penetration barriers are directly applicable to traditional mosquito adulticiding, which relies on dispersing fine droplets through the air to impinge upon and penetrate the cuticle of the target.

However, volatile pyrethroids diffuse as a gas, allowing passage through the thin, permeable endocuticle of the trachea (Sumita et al. 2016) and potentially avoiding some cuticle barriers.

Thus, one might expect greater penetration from volatile pyrethroids than more contact-active pyrethroids perhaps due to by-passed epicuticle penetration and reduced metabolism due to halogenated sites. Such reduced resistance to transfluthrin was observed in the metabolically- active FUMOZ-R strain of An. Funestus (Giles) (Horstmann and Sonneck, 2016), and the Puerto

Rico strain of Ae. aegypti, where in this strain resistance ratios to permethrin, deltamethrin, and transfluthrin were 112-, 650-, and 29-fold, respectively (Agramonte et al., 2017).

Spatial repellent tools are one of several potential delivery mechanism for volatile pyrethroid active ingredients. It has been shown that metofluthrin, via the OFF! Mosquito Lamp,

generates mosquito knockdown and mortality up to 15m downwind of the point source (Shen et al. 2017). This supports the possibility of transitioning these compounds into drift-dependent space sprays. It remains unclear what sub-lethal impacts might occur in vectors when the targets are exposed to pyrethroid vapors. Understanding sub-lethal effects can enhance understanding of the full spectrum of spatial repellent effects upon vector species, and lead to deployment strategies with greater flexibility and improved outcomes. For example, prallethrin has been used to successfully reduce the field abundance of Ae. albopictus when the chemical was delivered as an ultra-low volume contact spray (Farajollahi et al. 2012) The compound was described as causing a benign excitation that flushed mosquitoes out of harborage to allow for greater exposure and effective kill (Farajollahi et al. 2012). Transfluthrin, meperfluthrin, and metofluthrin could be substituted for prallethrin, while still being effective as a mosquito adulticide based on vapor bioassays and drift-enhanced range of point source treatments (Shen et al. 2017). Alternatively, they could be evaluated for use in mosquito abatement operations, such as through formulations that require less chemical output, or in strategies that circumvent barriers to effective treatment with the currently available products. The potential applications and strong candidacy of transfluthrin, meperfluthrin, and metofluthrin as vapor-active insecticides warrant further development for mosquito abatement.

2.5 Disclosures

This article was published in the Society of Chemical Industry journal of Pest

Management Science in 2018, volume 74, issue 12, pages 2699-2706. Funding for this research provided by the Florida Department of Agriculture and Consumer Services project 23583.

2.6 Tables

Table 2-1. Comparative LC50 values of five volatile pyrethroids, delivered as a vapor, to pyrethroid susceptible strains of Aedes albopictus and Ae. aegypti.

LC (± 95% FL)† 50 Exposure Pyrethroid Ae. albopictus Slope χ2 n Ae. aegypti Slope χ2 n 2h ‡ Transfluthrin 0.062 (0.054 – 0.071) 2.5±0.3 42.09 1040 0.066 (0.025 – 0.106) 2.1±0.2 339.86 1040 Meperfluthrin 0.045 (0.040 – 0.050) 1.9±0.2 26.71 720 0.033 (0.029 – 0.036) 2.1±0.1 11.12 720 Metofluthrin 0.68 (0.57 – 0.80) 1.9±0.2 87.43 720 0.28 (0.24 – 0.34) 1.4±0.1 27.86 1360 Prallethrin 1.1 (0.9 – 1.2) 2.4±0.2 33.74 960 0.68 (0.53 – 0.85) 1.9±0.1 99.56 960 Flumethrin 17 (15 – 22) 1.8±0.3 22.83 960 10 (5 – 23) 1.0±0.2 16.98 960 4h ‡ Transfluthrin 0.044 (0.035 – 0.052) 2.8±0.3 167.84 1040 0.068 (0.038 – 0.100) 2.6±0.2 312.45 1040 Meperfluthrin 0.035 (0.031 – 0.039) 1.7±0.1 25.51 720 0.027 (0.024 – 0.030) 1.9±0.1 9.75 720 Metofluthrin 0.55 (0.49 – 0.62) 3.3±0.4 43.46 720 0.16 (0.13 – 0.20) 1.5±0.1 26.45 1360 Prallethrin 0.45 (0.33 – 0.59) 1.6±0.1 59.43 960 0.44 (0.32 – 0.57) 1.8±0.1 126.52 960 Flumethrin 13 (11 – 16) 3.4±0.5 19.52 960 2.7 (1.6 – 4.7) 1.0±0.2 9.32 960 24h ‡ Transfluthrin 0.038 (0.032 – 0.044) 7.8±0.8 1.65 1040 0.007 (0.001 – 0.021) 3.8±0.4 28.14 1040 Meperfluthrin 0.027 (0.023 – 0.030) 1.8±0.1 14.36 720 0.022 (0.019 – 0.025) 1.8±0.1 10.74 720 Metofluthrin 0.41 (0.36 – 0.48) 2.5±0.2 58.53 720 0.06 (0.05 – 0.07) 1.4±0.1 25.06 1360 Prallethrin 0.28 (0.22 – 0.35) 1.8±0.2 32.87 960 0.36 (0.27 – 0.47) 2.4±0.2 106.05 960 Flumethrin 10 (8 – 11) 4.0±0.5 19.76 960 2.2 (1.3 – 3.8) 0.7±0.1 10.59 960 2h +, Transfluthrin 0.067 (0.064 – 0.070) 2.9±0.3 92.42 1040 0.064 (0.060 – 0.067) 7.5±0.6 15.3 1040 Recovery § Meperfluthrin 0.057 (0.051 – 0.064) 2.0±0.2 13.25 720 0.039 (0.035 – 0.043) 2.0±0.2 12.07 720 Metofluthrin 0.93 (0.80 – 1.12) 3.1±0.2 53.33 720 0.09 (0.07 – 0.10) 1.3±0.1 24.6 1360 Prallethrin 0.68 (0.58 – 0.79) 3.5±0.4 56.9 960 1.4 (1.1 – 1.8) 0.9±0.1 28.46 960 Flumethrin 13 (11 – 16) 1.9±0.2 36.1 960 7.5 (4.8 – 12.7) 0.6±0.1 8.19 960

† Values are LC50 with 95% fiducial limits (FL; lower FL, upper FL) shown in m/v (mass/volume; g/100 ml). Based on serial dilutions of compounds applied to 5 × 40 mm (200 mm2) filter paper strips in a 473.18 ml air space. Slopes are ± SEM. ‡ Mosquitoes were exposed continuously in the test container and mortality recorded at 2h, 4h, and 24h. § Recovery assays where mosquitoes were exposed for 2h and transferred to separate, clean containers to allow metabolic recovery. Mortality recorded 24h after initial exposure.

Table 2-2. Comparative LC50 values of five volatile pyrethroids, delivered as a vapor, to pyrethroid susceptible strains of Culex quinquefasciatus and Anopheles quadrimaculatus.

LC (± 95% FL)† 50 Exposure Pyrethroid Cx. quinquefasciatus Slope χ2 n An. quadrimaculatus Slope χ2 n

2h ‡ Transfluthrin 0.023 (0.017 – 0.028) 1.4±0.1 17.55 1040 0.031 (0.027 – 0.036) 1.9±0.1 26.66 1040 Meperfluthrin 0.033 (0.028 – 0.037) 1.7±0.1 10.78 720 0.022 (0.018 – 0.026) 1.5±0.1 7.65 720 Metofluthrin 0.47 (0.40 – 0.55) 1.7±0.1 20.67 1120 0.050 (0.043 – 0.060) 1.8±0.1 30.51 1040 Prallethrin 0.26 (0.21 – 0.32) 1.7±0.2 27.81 1120 0.48 (0.40 – 0.57) 1.7±0.1 35.74 1120 Flumethrin 4.1 (2.8 – 6.0) 0.8±0.1 17.26 960 49 (26 – 173) 1.0±0.1 30.58 960

4h ‡ Transfluthrin 0.016 (0.011 – 0.020) 1.7±0.1 21.67 1040 0.023 (0.019 – 0.026) 1.9±0.1 22.17 1040 Meperfluthrin 0.029 (0.025 – 0.033) 1.7±0.1 8.24 720 0.017 (0.014 – 0.021) 1.5±0.1 6.67 720 Metofluthrin 0.30 (0.25 – 0.36) 1.4±0.1 76.01 1120 0.040 (0.034 – 0.046) 1.8±0.1 22.71 1040 Prallethrin 0.20 (0.16 – 0.25) 2.0±0.2 18.68 1120 0.39 (0.33 – 0.46) 1.7±0.1 43.69 1120 Flumethrin 3.2 (2.2 – 5.0) 0.9±0.1 20.09 960 43 (25– 114) 0.9±0.1 18.28 960 24h ‡ Transfluthrin 0.013 (0.009 – 0.017) 1.7±0.1 30.63 1040 0.016 (0.013 – 0.018) 1.9±0.1 23.35 1040 Meperfluthrin 0.016 (0.013 – 0.020) 1.4±0.1 14.45 720 0.014 (0.011 – 0.016) 1.7±0.1 6.98 720 Metofluthrin 0.19 (0.15 – 0.24) 1.4±0.1 55.29 1120 0.029 (0.025 – 0.033) 2.1±0.1 29.17 1040 Prallethrin 0.12 (0.09 – 0.14) 1.3±0.1 18.97 1120 0.27 (0.22 – 0.33) 1.5±0.1 35.43 1120 Flumethrin 2.7 (1.7 – 4.2) 1.0±0.1 14.42 960 24 (15 – 61) 1.0±0.2 19.51 960 2h +, Transfluthrin 0.027 (0.021 – 0.033) 1.9±0.1 29.98 1040 0.053 (0.048 – 0.059) 2.4±0.2 12.97 1040 Recovery § Meperfluthrin 0.044 (0.040 – 0.049) 2.0±0.1 3.94 720 0.026 (0.022 – 0.030) 1.5±0.1 8.3 720 Metofluthrin 0.79 (0.50 – 2.04) 1.5±0.1 48.07 1120 0.038 (0.034 – 0.043) 2.3±0.1 37.36 1040 Prallethrin 0.34 (0.30 – 0.46) 2.2±0.2 26.56 1120 0.55 (0.45 – 0.68) 1.7±0.1 65.35 1120 Flumethrin 6.4 (4.5 – 9.3) 0.9±0.1 23.29 960 53 (29 – 145) 0.9±0.1 13.35 960

Table 2-3. Comparative LC90 values of five volatile pyrethroids, delivered as a vapor, to pyrethroid susceptible strains of Aedes albopictus and Ae. aegypti

LC (± 95% FL)† 90 Exposure Pyrethroid Ae. albopictus Slope χ2 n Ae. aegypti Slope χ2 n

2h ‡ Transfluthrin 0.22 (0.17 – 0.31) 2.5±0.3 42.09 1040 0.33 (0.17 – 6.69) 2.1±0.2 339.86 1040 Meperfluthrin 0.22 (0.17 – 0.30) 1.9±0.2 26.71 720 0.14 (0.12 – 0.17) 2.1±0.1 11.12 720 Metofluthrin 2.5 (1.9 – 3.7) 1.9±0.2 87.43 720 2.5 (1.8 – 3.6) 1.4±0.1 27.86 1360 Prallethrin 3.7 (2.9 – 5.0) 2.4±0.2 33.74 960 3.2 (2.3 – 5.3) 1.9±0.1 99.56 960 Flumethrin 97 (61 – 208) 1.8±0.3 22.83 960 1,829 (392 – 2,792) 1.0±0.2 16.98 960

4h ‡ Transfluthrin 0.13 (0.11 – 0.18) 2.8±0.3 167.84 1040 0.24 (0.15 – 1.68) 2.6±0.2 312.45 1040 Meperfluthrin 0.19 (0.15 – 0.26) 1.7±0.1 25.51 720 0.13 (0.11 – 0.16) 1.9±0.1 9.75 720 Metofluthrin 1.6 (1.3 – 2.1) 3.3±0.4 43.46 720 1.2 (0.8 – 1.9) 1.5±0.1 26.45 1360 Prallethrin 3.0 (2.0 – 5.3) 1.6±0.1 59.43 960 2.3 (1.6 – 3.8) 1.8±0.1 126.52 960 Flumethrin 57 (38 – 117) 3.4±0.5 19.52 960 271 (93 – 1,488) 1.0±0.2 9.32 960 24h ‡ Transfluthrin 0.09 (0.08 – 0.10) 7.8±0.8 1.65 1040 0.11 (0.07 – 0.25) 3.8±0.4 28.14 1040 Meperfluthrin 0.14 (0.11 – 0.18) 1.8±0.1 14.36 720 0.11 (0.09 – 0.14) 1.8±0.1 10.74 720 Metofluthrin 1.5 (1.2 – 2.0) 2.5±0.2 58.53 720 0.50 (0.37 – 0.74) 1.4±0.1 25.06 1360 Prallethrin 2.0 (1.4 – 2.9) 1.8±0.2 32.87 960 1.6 (1.1 – 2.5) 2.4±0.2 106.05 960 Flumethrin 32 (24 – 56) 4.0±0.5 19.76 960 233 (81 – 1,274) 0.7±0.1 10.59 960 2h +, Transfluthrin 0.10 (0.09 – 0.11) 2.9±0.3 92.42 1040 0.09 (0.09 – 0.10) 7.5±0.6 15.3 1040 Recovery § Meperfluthrin 0.26 (0.20 – 0.36) 2.0±0.2 13.25 720 0.17 (0.14 – 0.23) 2.0±0.2 12.07 720 Metofluthrin 2.4 (1.8 – 3.7) 3.1±0.2 53.33 720 0.80 (0.57 – 1.24) 1.3±0.1 24.6 1360 Prallethrin 1.6 (1.2 – 2.8) 3.5±0.4 56.9 960 34 (20 – 68) 0.9±0.1 28.46 960 Flumethrin 60 (41 – 113) 1.9±0.2 36.1 960 1,689 (528 – 9,406) 0.6±0.1 8.19 960

† Values are LC90 with 95% fiducial limits (FL; lower FL, upper FL) shown in m/v (mass/volume; g/100 ml). Based on serial dilutions of compounds applied to 5 × 40 mm (200 mm2) filter paper strips in a 473.18 ml air space. Slopes are ± SEM. ‡ Mosquitoes were exposed continuously in the test container and mortality recorded at 2h, 4h, and 24h. § Recovery assays where mosquitoes were exposed for 2h and transferred to separate, clean containers to allow metabolic recovery. Mortality recorded 24h after initial exposure.

Table 2-4. Comparative LC90 values of five volatile pyrethroids, delivered as a vapor, to pyrethroid susceptible strains of Culex quinquefasciatus and Anopheles quadrimaculatus

† LC90 (± 95% FL) Exposure Pyrethroid Cx. quinquefasciatus Slope χ2 n An. quadrimaculatus Slope χ2 n

2h ‡ Transfluthrin 0.18 (0.13 – 0.29) 1.4±0.1 17.55 1040 0.15 (0.12 – 0.20) 1.9±0.1 26.66 1040 Meperfluthrin 0.18 (0.14 – 0.25) 1.7±0.1 10.78 720 0.17 (0.13 – 0.24) 1.5±0.1 7.65 720 Metofluthrin 3.3 (2.5 – 4.8) 1.7±0.1 20.67 1120 0.26 (0.19 – 0.40) 1.8±0.1 30.51 1040 Prallethrin 2.2 (1.6 – 3.4) 1.7±0.2 27.81 1120 2.6 (1.9 – 4.0) 1.7±0.1 35.74 1120 Flumethrin 145 (72 – 399) 0.8±0.1 17.26 960 1,029 (252 – 2,446) 1.0±0.1 30.58 960

4h ‡ Transfluthrin 0.09 (0.07 – 0.14) 1.7±0.1 21.67 1040 0.11 (0.09 – 0.14) 1.9±0.1 22.17 1040 Meperfluthrin 0.16 (0.13 – 0.22) 1.7±0.1 8.24 720 0.13 (0.10 – 0.17) 1.5±0.1 6.67 720 Metofluthrin 2.2 (1.6 – 3.2) 1.4±0.1 76.01 1120 0.21 (0.16 – 0.31) 1.8±0.1 22.71 1040 Prallethrin 1.5 (1.1 – 2.1) 2.0±0.2 18.68 1120 2.3 (1.7 – 3.3) 1.7±0.1 43.69 1120 Flumethrin 129 (61 – 377) 0.9±0.1 20.09 960 1,467 (391 – 1,733) 0.9±0.1 18.28 960

24h ‡ Transfluthrin 0.08 (0.06 – 0.12) 1.7±0.1 30.63 1040 0.07 (0.06 – 0.09) 1.9±0.1 23.35 1040 Meperfluthrin 0.14 (0.10 – 0.19) 1.4±0.1 14.45 720 0.08 (0.07 – 0.10) 1.7±0.1 6.98 720 Metofluthrin 1.7 (1.3 – 2.5) 1.4±0.1 55.29 1120 0.12 (0.10 – 0.16) 2.1±0.1 29.17 1040 Prallethrin 1.2 (1.0 – 1.8) 1.3±0.1 18.97 1120 1.9 (1.4 – 2.8) 1.5±0.1 35.43 1120 Flumethrin 119 (55 – 376) 1.0±0.1 14.42 960 686 (176 – 2,326) 1.0±0.2 19.51 960 2h +, Transfluthrin 0.13 (0.10 – 0.20) 1.9±0.1 29.98 1040 0.18 (0.15 – 0.23) 2.4±0.2 12.97 1040 Recovery § Meperfluthrin 0.20 (0.16 – 0.26) 2.0±0.1 3.94 720 0.19 (0.15 – 0.29) 1.5±0.1 8.3 720 Metofluthrin 3.5 (1.6 – 8.1) 1.5±0.1 48.07 1120 0.14 (0.11 – 0.17) 2.3±0.1 37.36 1040 Prallethrin 1.6 (1.1 – 2.5) 2.2±0.2 26.56 1120 3.0 (2.1 – 5.0) 1.7±0.1 65.35 1120 Flumethrin 247 (119 – 700) 0.9±0.1 23.29 960 2,006 (513 – 2,457) 0.9±0.1 13.35 960

2.7 Figures

Figure 2-1. Simplified vapor bioassay exposure chamber. Translucent polyethylene test containers with a volume of 473 ml. Snap-lids were modified with a 20-mm opening to allow admission of 20 female, non-blood-fed, 5–7d old mosquitoes. Treated filter paper strips were contained within a mesh bag suspended from the opening in order to allow passage of vapors while excluding direct contact. Container openings were sealed during testing.

Figure 2-2. Perfumery strip saturation method. Chemical solution was applied to a Whatman No.1 filter paper, which was cut into strips with dimensions of 5 mm × 40 mm, pleated every 5-mm in length. Applications were made by using a 20-μl pipette fitted with a filter tip to administer 40 μl of solution in two passes. Aliquots were kept in amber borosilicate vials to protect the chemical integrity.

Figure 2-3. Toxicant Relationship Diagram: Generalized relative toxicity of flumethrin, prallethrin, metofluthrin, transfluthrin, and meperfluthrin against mosquitoes; displayed with structures.

CHAPTER 3 SUB-LETHAL EFFECTS OF TRANSFLUTHRIN ON DOMESTIC MOSQUITO FECUNDITY AND OVIPOSITION BEHAVIOR

3.1 Introduction

The emergence of tropical pathogens, particularly Zika virus and its link to congenital birth defects (Chouin-Carneiro et al. 2015, Guo 2016), renewed emphasis on domestic Aedes aegypti (L.) and Aedes albopictus (Skuse) as urban disease vectors in the United States following local transmission of Zika virus in Florida (CDC 2016). These mosquito species disperse eggs across many natural and artificial containers holding small quantities of water that permeate urban landscapes (Devine et al. 2009). These simplified aquatic ecosystems virtually eliminate larval predation pressure, resulting in source reduction being the only long-term pressure to reduce densities of larval habitat used by container-inhabiting mosquitoes. In some environments, targeting key oviposition sites with source reduction has been effective in reducing container-inhabiting mosquito density (Maciel-de-Freitas and Lourenço-de-Oliveira

2011, Faraji and Unlu 2016). However, Ae. aegypti and Ae. albopictus bet-hedge by using a skip- oviposition strategy (Fay and Perry 1965, Davis et al. 2016), whereby eggs are distributed across multiple developmental sites per clutch. Because of skip-oviposition, it is difficult to eliminate all of the numerous mosquito oviposition sites from a peridomestic landscape. Oviposition sites also are often cryptic and difficult to access, further limiting the effectiveness of source reduction and cultural control (Devine et al. 2009, Suman et al. 2014). This places reliance on continuing to conduct chemical adulticiding, such as through ultra-low volume space sprays or outdoor residual treatments, which have variable rates of success (Farajollahi et al. 2012, Faraji and Unlu

2016). Consequently, personal repellents are an essential part of integrated vector management because it allows an extra layer of prevention when other treatments fail (CDC 2016).

In the current market, topical repellents form the core of personal protection guidelines against mosquito vectors (CDC 2016). However, many people do not use topical repellents because they find them inconvenient to apply or unpleasant in smell or feel when applied. Spatial repellent devices, such as ambient emanators, mosquito coils, vaporizer mats, and liquid vaporizers are an alternative to topical repellents that are frequently chosen by consumers (Xue et al. 2012b). Spatial repellents allow the dissemination of active ingredients on small scales, often serving as personal protection devices (Ritchie and Devine 2013, Bibbs and Xue 2015,

Bibbs et al. 2015). When such devices emit volatile pyrethroids, these tools can both reduce mosquito contact with humans (Xue et al. 2012b) and kill mosquitoes outright (Bibbs and Xue

2015, Bibbs et al. 2015). This is due to the primary mode of action of pyrethroids, in which the active ingredient binds to the voltage gated sodium channel (VGSC) as the main mode of toxicity. As reviewed in other work (Achee et al. 2012a, Bibbs and Kaufman 2017), volatile pyrethroids can bind to the VGSC after inhalation. This property of inhalation is a clear benefit of some active ingredients, such as metofluthrin or transfluthrin, that lead to various acute and sub-acute outcomes. Because of the combined benefits of repellency and mortality, volatile pyrethroids have been advocated as a tool for urban vector management (Ritchie and Devine

2013, Bibbs and Xue 2015, Bibbs et al. 2015).

While volatile pyrethroids are repellents that also can cause mortality, mosquitoes have many opportunities to survive exposure. Mosquitoes may escape the area before acquiring a lethal dose or strong air movement may disperse the vapors. This range of exposures potentially produces a suite of sub-lethal outcomes depending on what point the mosquito escapes exposure

(Achee et al. 2012a, Bibbs and Kaufman 2017). Sub-lethal impairment of mosquito reproduction by spatial repellents is relevant for container-inhabiting mosquitoes because they live in close

association with humans and have relatively short range oviposition site use (Davis et al. 2016).

This allows spatial repellents to impact container-inhabiting mosquitoes when host-seeking, gravid, or ready for oviposition. Some mosquitoes survive and escape exposure to spatial repellents, while others show signs of toxicity after escaping, summarized in Achee et al. (2012a) and Bibbs and Kaufman (2017). Yet, the degree to which sub-lethal exposure to volatile pyrethroids in spatial repellents can negatively affect downstream fecundity or oviposition behaviors is unknown.

In related work, a wide array of experimental and commercial skin repellent compounds have been shown to deter Ae. albopictus from ovipositing in containers fitted with a repellent treated barrier (Bar-Zeev and Ben-Tamar 1968) or repellent contaminated water (Xue et al. 2003;

2006). Furthermore, Xue et al. (2004) found that when oviposition sites were contaminated with

DEET, gravid mosquitoes retained eggs for as much as 3 weeks after the exposure. Furthermore, the viability of these retained eggs decreased as more time elapsed before mosquitoes were able to successfully deposit their eggs (Xue et al. 2004). Choi et al. (2016) exposed gravid Ae. aegypti to a volatile pyrethroid, transfluthrin, and found that bacteria-baited oviposition cups were twice as attractive to treated mosquitoes and that treated mosquitoes had less overall dispersion of eggs despite being offered multiple containers. Whether the observed changes in oviposition were due to disrupted skip-oviposition patterns is not discussed by that study. However, the growing observations of sub-lethal effects should be applied to the peridomestic ecology of Ae. aegypti and Ae. albopictus to quantify how spatial repellents affect mosquitoes surviving exposure. The effects of exposure to sub-lethal concentrations of transfluthrin vapors were quantified for Ae. aegypti and Ae. albopictus fecundity and oviposition behavior to test the extent to which these chemicals damage mosquito reproductive performance. It was predicted that exposing Ae.

aegypti and Ae. albopictus to sub-lethal concentrations of volatile pyrethroid vapors would reduce mosquito fecundity in a subsequent oviposition attempt. Additionally, exposure to sub- lethal concentrations of volatile pyrethroid vapors were predicted to decrease the occurrence of skip-oviposition behavior in both species.

3.2 Materials and Methods

3.2.1 Insect Rearing

Pyrethroid susceptible, 1952 Orlando, FL strain Aedes aegypti and 1992 Gainesville, FL strain Aedes albopictus were acquired from the United States Department of Agriculture,

Agricultural Research Service, Center for Medical, Agricultural, and Veterinary Entomology

(USDA-ARS-CMAVE) in Gainesville, Florida. Colonies of the susceptible strain were not exposed to insecticides prior to evaluation and were not supplemented with wild-type introductions. Mosquitoes were kept at 26 ± 1°C, 85 ± 5% relative humidity (RH), with a 14:10

(L:D) photoperiod. Batches of 2,000 eggs were placed in larval pans containing 2,500 ml of reverse osmosis (RO) water. Hatched larvae were fed 1–3 g of liver and yeast mixture at a 3:2 ratio ad libitum in a 50-ml suspension. Maturing adult mosquitoes were kept in 30 × 30 × 30 cm flight cages. Flight cages also were supplied with 10% sucrose solution and RO water.

Mosquitoes used at the start of experimentation were non-blood-fed, 5–7d old females that were given the opportunity to mate without blood feeding.

3.2.2 Bioassay Design

Test cages were derived from single-use 473 ml clear polypropylene snap-lid cups.

Container lids were modified to have a central 20-mm opening through which 20 female mosquitoes were aspirated into the container. Filter paper was cut into 5-mm widths and 40-mm lengths and pleated every 5 mm. Technical grade transfluthrin was serially diluted in acetone to create concentrations meeting LC10, LC20, and LC30 predicted dose responses using 24h mortality

data from Bibbs et al. (2018b). The predicted concentrations from Bibbs et al. (2018b) were derived from acute toxicity dose responses after 2h exposures, but the current concentrations were selected with the intent of avoiding acute mortality while still allowing signs of toxicity to be detected several days later. These concentrations in solution were 0.009% for LC10, 0.016% for LC20, and 0.026% for LC30 against Ae. aegypti and 0.012% for LC10, 0.020% for LC20, and

0.029% for LC30 against Ae. albopictus.

Paper strips were then treated with 40 μl of a transfluthrin solution. Treated strips were air dried for 6min and then transferred into a mesh bag that was suspended within the test cage through the hole in the lid. The hole was sealed to prevent vapors from escaping during tests.

Controls used paper strips treated with only acetone. Test cages were placed in an illuminated incubator maintained at 26 ± 1°C, 85 ± 5% RH for a 2h exposure period. The 2h exposure was selected to adhere to the protocol of Bibbs et al. (2018b) and validate that the predicted concentrations abide the LC10, LC20, and LC30 acute toxicity outcomes.

Following the 2h transfluthrin exposure, mosquitoes were removed into a clean container of the same design. Upon transfer to clean containment, both treatment and unexposed control cohorts were offered 100ml of non-citrated bovine blood in an unsalted sausage casing after warming to 36±0.5°C using a hot water bath. Freshly warmed blood meals were supplied for 2h per cohort, with blood replaced each hour. Post-blood-meal, mosquitoes lacking a fully engorged, red abdomen, as compared to control mosquitoes, were removed from the assay. The remaining mosquitoes were allowed 72h post blood meal to become gravid while held in the rearing conditions. Gravid mosquitoes for each cohort were confirmed with visual inspection of the abdominal membrane and transferred individually to polypropylene flight cages measuring

30 x 30 x 30cm for oviposition (Bugdorm I # DP1000, MegaView Science Co. Ltd., Taiwan).

Groups of five flight cages, each with one mosquito per cage, were used for LC10, LC20, and

LC30 treatments, and another group of five cages were used for control, totaling 20 cages per replicate. Each flight cage contained a source of 10% sucrose solution and six black, 118ml oviposition cups (P400BLK, Dart Container Corporation, Mason, MI). Each cup was filled with

50ml of RO water and a 5 x 8 cm strip of oviposition paper (#6,512,981,311, Anchor Paper Co.,

Saint Paul, MN). Gravid mosquitoes were allowed 72h to oviposit across the array of cups. The parameters recorded from the oviposition bioassay included how many eggs were deposited, how many cups contained eggs, and the living or deceased status of the treated female mosquito. The flight-oviposition cages were randomly assigned a 0°, 90°, 180°, or 270° orientation within an insectary at the beginning of the study to reduce the light effects and orientation biases.

Once oviposition bioassays were concluded, adult mosquitoes were immediately killed and dissected in 70% ethyl alcohol under a light microscope to examine egg retention.

Dispersion of eggs laid across cups and quantity of eggs for each cup were recorded upon removal of adults for dissection. To facilitate embyronation of the eggs, oviposition papers were stored in rearing conditions and loosely enveloped in wax paper for a 24h damp drying period.

Eggs were inspected for any deformities following damp drying and then submerged in 118-ml cups filled with 50 ml of RO water. Larvae were reared out post-hatch with larval food provided ad libitum until all larvae pupated or perished. The number of larvae that hatched and subsequent natal mortality were recorded daily over the rearing period. Unhatched eggs were examined for deformities again once larval rearing from the selected paper was concluded. Eggs were denoted as non-viable if at either examination the chorion was collapsed. The progression required to complete a repetition of this bioassay appears in Fig. 3-1. This design was replicated across six different cohorts of mosquitoes for each species.

3.2.3 Data Analysis

Kruskal-Wallis analysis of variance and post hoc Steel-Dwass tests were performed on averages of eggs collected, number of cups used by females, number of larvae successfully hatched and reared, number of eggs that failed to embryonate, and the number of eggs retained in dissected adult females after the bioassay to compare effects across treatments. Delayed mortality of adult females for each treatment, whereby Ae. aegypti that were exposed to transfluthrin prior to blood feeding were discovered in a prone position rigidly immobilized, during the oviposition bioassay, but not within 24h of transfluthrin exposure, was compared to the controls using a Chi-square analysis. For Ae. albopictus, a melanization effect observed in retained eggs also was analyzed for each treatment using a Chi-square analysis. The severity of oocyte melanization in each female was qualitatively placed into one of five groups: 0% of oocytes melanized, 1–25% of oocytes melanized, 26–50% of oocytes melanized, 51–75% of oocytes melanized, and 76–100% of visible oocytes melanized and proportion of each group were compared among treatments. Statistical procedures were repeated for all concentrations and species in JMP 13.1.0 (SAS Institute, Inc., Cary, NC).

3.3 Results

3.3.1 Aedes aegypti

Exposure of female Ae. aegypti to sub-lethal concentrations of transfluthrin reduced the

2 dispersion of eggs across the oviposition arena (Fig. 2-2, χ (3) = 66.2, p < 0.0001). All transfluthrin exposure groups oviposited in fewer cups than the unexposed controls, but there was no differences in egg dispersion between the transfluthrin exposure groups (mean ranks from Kruskal-Wallis tests were 103.2 for controls, 46.6 for LC10, 46.3 for LC20, and 45.9 for

LC30). Aedes aegypti from control groups used all six oviposition containers in 83.3% of the bioassays (Fig 3-2). Mosquitoes oviposited in 1–2 containers an average 90.0% and 86.7% of the

bioassays after the LC10 and LC20 treatments, respectively. Mosquitoes exposed to the LC30 treatment oviposited in 0–2 containers on average in 86.7% of the bioassays. There was also a significant reduction in the total number of eggs oviposited by Ae. aegypti females following

2 sub-lethal transfluthrin exposure compared to unexposed controls (Fig. 3-3, χ (3) = 187.6, p <

0.0001, n=30), but there were no differences in total number of eggs oviposited among our volatile-pyrethorid exposure groups (mean ranks from Kruskal-Wallis tests were 524.5 for controls, 308.4 for LC10, 306.1 for LC20, and 303.0 for LC30).

Female Ae. aegypti exposed to transfluthrin also experienced reduced egg viability. This was evident because many eggs were collapsed and few hatched in the sub-lethal exposure groups (Fig. 3-4). Collapsed eggs tended to occur in clusters, but were sometimes mixed with viable eggs (Fig. 3-4). In ~30% of samples, the entire collection of eggs on an oviposition paper were collapsed by the end of the larval hatching period. The LC10 exposure group did not significantly differ in the number of collapsed eggs from the controls. However, the LC20 and

LC30 exposures had significantly higher numbers of collapsed eggs than those observed in either

2 the LC10 or control treatments (Fig. 3-3, χ (3) = 76.0, p < 0.0001, n=30, mean ranks from

Kruskal-Wallis tests were 320.0 for control, 327.2 for LC10, 393.6 for LC20, and 401.1 for LC30).

Eggs that were not collapsed were considered potentially viable, therefore collapsed eggs were removed from the total and only the remaining eggs were hatched out. Of these, larval hatch was not significantly different across treatments and controls, but a weak visual trend indicated a negative correlation of hatch rate with exposures. In rearing the viable eggs, there was no difference in the survivorship to adulthood of hatched larvae between the control, LC10, and LC20 treatments. There was a weak trend suggesting that survivorship decreased 2% in

2 larvae hatching from the eggs laid by LC30-exposed female mosquitoes (χ (3) = 11.0, p < 0.0116),

with a mean rank of 373.6 for control, 368.5 for LC10, 351.8 for LC20, and 348.1 for LC30. There were no effects on pupation, adult emergence, or subsequent blood feeding and oviposition in the

F1 generation.

There was significant delayed toxicity at the higher exposure levels where Ae. aegypti that were exposed to transfluthrin prior to blood feeding were discovered dead six days after exposure when concluding oviposition bioassays. The LC20 exposure generated 36.7% delayed toxicity in adult females during the oviposition bioassay, and the LC30 exposure generated 60.0%

2 delayed toxicity (χ (3) = 52.9, p < 0.0001). No delayed toxicity was observed in LC10 or the controls. Even though they may have died soon after, all females that died during the oviposition bioassay survived long enough to lay eggs in the provided arena. Both living and dead mosquitoes were dissected after the oviposition bioassay to determine whether they retained any matured eggs in their reproductive tracts or whether all matured eggs were laid. During dissections, control mosquitoes did not retain any mature eggs, while the egg retention in the

LC10, LC20, and LC30 cohorts ranged from 0–3, 0–28, and 0–27 eggs per female, respectively.

2 Females in the LC20 group retained more mature eggs than controls (Fig. 3-3, χ (3) = 61.5, p <

0.0001, mean rank of 39.5 for control, 48.4 for LC10, 95.3 for LC20, and 58.8 for LC30). Although there was a visual trend towards egg retention in the LC30 group, this trend was not statistically detectably different from the control. Overall, when we combine each of the facets of reduced reproductive performance that we quantified in Ae. aegypti, mosquitoes exposed to the LC30 dose of transfluthrin vapors had ~70% reduction in viable eggs.

3.3.2 Aedes albopictus

Aedes albopictus also oviposited across significantly fewer sites than control mosquitoes

2 when exposed to any of the sub-lethal concentrations of transfluthrin (Fig. 3-2, χ (3) = 67.6, p <

0.0001, n=30, Kruskal-Wallis mean ranks of 104.4 for control, 50.5 for LC10, 44.3 for LC20, and

42.8 for LC30). As above, there significant reduction in the total number of eggs deposited by mosquitoes exposed to any of the transfluthrin concentrations compared to controls (Fig. 3-3,

2 χ (3) = 242.7, p < 0.0001, n=30, mean ranks of 550.5 for control, 315.0 for LC10, 286.1 for LC20, and 290.4 for LC30).

In contrast to Ae. aegypti, less than 1% of Ae. albopictus eggs were collapsed across all treatments and controls. Larval hatch was 98–100% successful for the control and all transfluthrin treatments, and larval survival after hatching was not significantly different across transfluthrin treatments and the control. Unlike Ae. aegypti, there was no delayed toxicity observed in adult Ae. albopictus as a result of exposure to transfluthrin vapors. When female mosquitoes were dissected to determine whether all matured eggs were laid, we found that Ae. albopictus controls did not retain any mature eggs, but the LC10 group collectively retained more mature eggs than either the control or higher-concentration transfluthrin treatments (Fig. 3-3,

2 Fig. 3-5, χ (3) = 54.8, p < 0.0001, n=30, mean ranks of 32.0 for control, 90.3 for LC10, 50.2 for

LC20, and 69.5 for LC30). Females in the LC30 treatment retained significantly more eggs than the control (Z = 5.5, p < 0.0001) and LC10 (Z = −3.1, p < 0.0109) groups, but were not statistically different from the LC20 group.

During dissections, we found that in the LC30 treatment group 70% of the females contained a proportion of oocytes retained in the reproductive tract that were melanized (Fig. 3-

5e, 5f), but neither the control nor other sub-lethal exposure groups had melanized oocytes (Fig.

2 5a–d, χ (12) = 74.6, p < 0.0001). The ratio of melanized eggs with respect to retained and oviposited eggs are displayed in Fig. 3. If melanized eggs were not viable, the combined effects that reduce reproductive performance in Ae. albopictus represent ~65% total reduction of viable eggs after exposure to the LC30 of transfluthrin vapors.

3.4 Discussion

Exposure to sub-lethal concentrations of transfluthrin vapors reduced female reproduction, including fecundity, fertility, and the dispersion of eggs across potential oviposition sites in both Aedes species. The breadth of impacts to reproductive performance add to the spectrum of outcomes against mosquitoes when using spatial repellents. Skip-oviposition in particular is a critical behavior to interrupt because untreated Ae. aegypti and Ae. albopictus consistently spread eggs to 4–6 containers per gonotrophic cycle, as seen in the control mosquitoes (Oliva et al. 2014, Fonesca et al. 2015, Davis et al. 2016, Santos de Abreu et al.

2016). However, skip-oviposition is susceptible to external pressures, it wanes if there are limited oviposition sites available (Santos de Abreu et al. 2015), during seasonal changes

(Fonseca et al. 2015), and in certain geographic localities (Harrington and Edman 2001). Source reduction ideally should eliminate skip-oviposition behavior by removing options for oviposition

(Santos de Abreu et al. 2015). In practice, it has become evident that container-inhabiting mosquitoes find oviposition sites despite source reduction efforts, thereby confounding the sustainability of source reduction as a mosquito abatement strategy (Faraji and Unlu 2016).

Volatile pyrethroids may provide an additional external pressure needed to manipulate skip- oviposition behavior in Ae. aegypti and Ae. albopictus to facilitate source reduction impacts within an integrated approach.

Our results represent both a dramatic reduction of viable eggs and a favorable reduction of skip-oviposition behavior by both Ae. aegypti and Ae. albopictus upon exposure to transfluthrin. The changes in oviposition behavior in our work may be a result of behavioral modification by neuronal interference, since transfluthrin binds to the VGSC. A complimentary effect was discussed by Choi et al. (2016) where Ae. aegypti displayed increased attraction to oviposition sites after sub-lethal exposure to transfluthrin. When paired with the reduced

reproductive performance observed in the present study, spatial repellents appear to stimulate container-inhabiting mosquitoes to urgently oviposit in nearby containers. Therefore, volatile pyrethroids can synergize with source reduction programs by reducing the labor necessary to target key oviposition sites (Maciel-de-Freitas and Lourenço-de-Oliveira 2011). Additionally, there are other circumstances where mosquitoes may get unintentional sub-lethal exposures. It has suggested that the time it takes for the vapors to penetrate into surroundings can lead to reduced exposure of target mosquitoes (Buhagiar et al. 2017a; 2017b). Our findings also support that mosquitoes that may escape spatial repellents can still incur various side-effects. A mosquito that approaches hosts protected by spatial repellents on multiple occasions may even experience several repeated sub-lethal exposures through its lifetime.

However, the current delivery methods for vapor-active pyrethroids as spatial repellents are restricted to managing small areas. With current delivery methods protecting large areas would be as labor intensive and costly as source reduction (Maciel-de-Freitas and Lourenço-de-

Oliveira 2011, Faraji and Unlu 2016). Given the ecology of domestic mosquitoes and the growing spectrum of benefits that volatile pyrethroids have against container-inhabiting mosquitoes, volatile pyrethroids should be developed into tools that are more capable of addressing the needs of mosquito abatement programs. Recent field studies support the idea that vapor-active pyrethroids can be deployed in ultra-low volume sprays to suppress domestic mosquitoes (Farajollahi et al. 2012, Unlu et al. 2014). Vapor-active pyrethroids also may be compatible with other delivery formats that are useful for integrated vector management.

Although our results are proof of concept that spatial repellents can harm mosquito reproduction 6d after exposure, many details are not well understood and will require further study. When breaking apart the fertility measurements, Ae. albopictus had an unusual pattern

whereby all of the treatments had lower egg viability than the control, but there was an increasing pattern of egg viability with higher transfluthrin concentration. A possible explanation is hormesis (Cutler 2013), a well-documented occurrence in which an organism (e.g., insects) experiences sub-lethal exposure to a stressor (e.g., toxicant or xenobiotic) and correspondingly displays increased fitness (e.g., egg production) at low doses while being inhibited at high doses

(Antonio et al. 2009, Bong et al. 2017). Hormesis is an effect observed in both field and laboratory test groups, and current evidence asserts that pesticide hormesis does not significantly differ in magnitude in laboratory versus field colonies for Ae. aegypti (Bong et al. 2017).

Unfortunately, our range of concentrations was not broad enough to observe a drop in the effect after a peak, so we cannot confirm what caused the pattern in viability or what it means for the animal. The effect was minimal for overall reproduction and the net effect still resulted in less oviposition compared to the control group. In contrast, the oocyte melanization observed in Ae. albopictus could be explained as an immune-system cost of surviving exposure. Cellular responses to stress, such as upregulation of detoxification enzymes or immune responses, have been shown to trigger melanization in the ovary and follicular apoptosis in Anopheles gambiae

Giles (Ahmed and Hurd 2006). In one Ae. albopictus female with 100% of the visible oocytes melanized, bacterial decomposition was visible in one of the ovaries of the otherwise surviving female mosquito. Stress responses could explain the observed melanization and tissue decay, reinforcing that sub-lethal exposures to volatile pyrethroids could have substantial implications for mosquito populations.

There are many potential directions for continuing work on sub-lethal effects caused by volatile pyrethroids. Although we monitored for mortality and reproductive performance for one week after exposure and blood feeding, there may be long-term effects of sub-lethal exposures

beyond the first gonotrophic cycle that should be examined. Exposure concentrations also can be controlled by time, and has been recognized as a relevant limiter for volatile pyrethroids when the vapors must travel and penetrate into areas of interest (Buhagiar et al. 2017a; 2017b). In an effort to make the findings of our proof of concept study more relevant to field conditions, a valuable next step would be to test short exposure durations. Particularly with respect to the fact mosquitoes may only briefly encounter the toxicant when approaching hosts. Regardless, this study demonstrated that even despite hormetic gains in one facet of reproduction, Ae. aegypti and

Ae. albopictus experienced reduced overall egg yield, viability, and skip-oviposition behavior following exposure to transfluthrin at a low concentration.

3.5 Disclosures

This article was published in the Springer Nature open access journal of Parasites and

Vectors in 2018, volume 11, page 486. Funding for this research was provided in part by the

Florida Department of Agriculture and Consumer Services: Florida Coordinating Council on

Mosquito Control research subcommittee project 23583. Publication of this article was funded by the Anastasia Mosquito Control District of St. Johns County.

3.6 Figures

Figure 3-1. Overall design of the experiment with time estimates showing the duration of steps.

Figure 3-2. Post transfluthrin vapor exposure decrease in egg dispersion across containers. Cluster graphs representing the mean percentage of adult Aedes aegypti (L.) (n=30) and Ae. albopictus (Skuse) (n=30) ovipositing in 0, 1, 2, 3, 4, 5, or 6 cups following exposure to three sub-lethal concentrations of transfluthrin (LC10, LC20, LC30). Figures shown with standard error of the mean as I-bars.

Figure 3-3. Post transfluthrin vapor exposure reduction in fecundity. The LC10, LC20, and LC30 of transfluthrin were tested on Aedes aegypti (L.) (n = 30) and Ae. albopictus (Skuse) (n = 30). Compound bar graphs represent the counts of collapsed eggs (Ae. aegypti), viable eggs, eggs retained in the parent female, and melanization of retained eggs (Ae. albopictus) following oviposition in a six-cup arena. The sums of: Collapsed or Melanized + Viable + Retained = total egg production, Collapsed + Viable = total eggs oviposited, Viable + Retained = yield that could recruit to the next generation. Figures are shown with 95% confidence as I-bars. Tables beneath graph detail range and mean counts per ovicup in the bioassay as ‘Viable [˗Collapsed].’

Figure 3-4. Multi-plate figure showing Aedes aegypti (L.) eggs following exposure of the parent female mosquito to sub-lethal concentrations of transfluthrin. A: Seed germination paper removed from an LC10 treatment oviposition bioassay cup, following a 24h damp dry, and a full drying period. Solid line encircles embryonated, viable eggs. Dashed line encircles non-viable eggs with a collapsed chorion. B: Cluster of collapsed eggs from an LC20 treatment oviposition bioassay, following a 24h damp dry period, but not fully dried. C: A cluster of eggs upon immediate removal from an LC30 treatment oviposition bioassay. D: The same cluster of eggs, collapsed, following a 24h damp dry and full drying period.

Figure 3-5. Multi-plate figure showing dissected female Aedes albopictus (Skuse) reproductive tracts after exposure to sub-lethal concentrations of transfluthrin and upon conclusion of oviposition bioassays seven days post-exposure. A: Control mosquito showing cleared oviducts and no late stage development oocytes. B: A reproductive tract following LC10 treatment with 10 or fewer retained eggs. C: A reproductive tract following LC20 treatment with 20 or fewer retained eggs. D: A reproductive tract following LC30 treatment with extreme egg retention. E: A reproductive tract following LC30 treatment showing partial melanization of retained eggs. F: A reproductive tract following LC30 treatment showing extreme melanization and decay.

CHAPTER 4 SUB-LETHAL EFFECTS OF VOLATILE PYRETHROIDS ON RESISTANT AND FIELD STRAINS OF AEDES AEGYPTI AFTER BRIEF EXPOSURE DURATIONS

4.1 Introduction

Aedes aegypti (L.) is a peridomestic and urban vector of dengue, chikungunya, Zika, and yellow fever viruses (Oliosi et al. 2018). This mosquito has opportunistic feeding patterns (Smith et al. 2018) and is an aggressive, day biting, anthropophilic pest that lays eggs in containerized impoundments of water, both natural and artificial (Devine et al. 2009). Integrated mosquito management increasingly is incorporating recommendations for spatial repellents when constituents are at risk of Ae. aegypti (Devine et al. 2009, Bibbs and Xue 2015). Among spatial repellents, volatile pyrethroids are active ingredients with the largest research base (Achee et al.

2012a, Bibbs and Kaufman 2017), and include certain active ingredients, such as metofluthrin and transfluthrin.

Recent work supports the toxicity of volatile pyrethroids, particularly those with fluorinated alcohol rings such as transfluthrin, meperfluthrin, and metofluthrin, when applied in a vapor phase (Sugano and Ishiwatari 2012, Manda et al. 2013, Ritchie and Devine 2013, Bibbs and Xue 2015, Buhagiar et al. 2017a; 2017b, Shen et al. 2017, Bibbs et al. 2018b). A broad spectrum of spatial repellent toxicological effects have been summarized in Achee et al. (2012a) and Bibbs and Kaufman (2017), yet sub-lethal effects are poorly characterized for mosquitoes assumed to escape a lethal exposure. Prior work exposed both Ae. aegypti and Ae. albopictus

Skuse to sub-lethal concentrations of transfluthrin vapors. Treated mosquitoes displayed altered oviposition behavior, oviposited fewer total eggs, and the eggs showed reduced viability (Bibbs et al. 2018a). Despite the prior results, concerns of whether pyrethroid resistant or wild-type mosquitoes are susceptible to the sub-lethal oviposition effects need to be assessed before recommending spatial repellents as an oviposition deterrent in operational mosquito control.

Pyrethroid resistance has been increasingly identified in Ae. aegypti populations across its introduced range (Amelia-Yap et al. 2018). A natural counter argument to incorporating spatial repellents into operational mosquito control to target oviposition is whether sub-lethal effects are relevant in resistant populations. For example, pyrethroids, including those used in spatial repellents, elicit repellency through excitation and sub-lethal neurological impairment (Manda et al. 2013). It is presumable that resistance mechanisms, such as knockdown resistance (kdr), could reduce the efficacy of sub-lethal effects following exposure to volatile pyrethroids.

However, pyrethroid tolerance or resistance does not necessarily prevent repellency (Bowman et al. 2018) and contact irritancy (Grieco et al. 2007, Achee et al. 2009) by pyrethroids, and in fact the repellent responses appear independent of the toxicity (Achee et al. 2009). Despite finding susceptibility of resistant phenotypes to the repellent action of pyrethroids, it is unknown whether the aforementioned sub-lethal effects are replicable in a field strain that experiences regular pressure from exposure to multiple types of pyrethroid.

Another critical matter is the context in which mosquitoes may be exposed to sub-lethal concentrations of spatial repellents. Prior work has advocated for incorporating spatial repellents into integrated vector management as a compliment to source reduction (Bibbs et al. 2018a).

However, the more realistic setting is that sub-lethal effects will occur when spatial repellents are employed as a personal protection aid. For example, spatially repellent vapors of metofluthrin prevent Anopheles (Kawada et al. 2008), Aedes (Kawada et al. 2006), and Culex (Kawada et al.

2005) vectors from biting people in a variety of settings. Yet, it is also known that turbulent air in the vicinity of a volatile emission may disperse vapors to a degree that prevents rapid knockdown and mortality, instead leading to drift of volatile pyrethroid plumes into adjacent or downwind areas (Shen et al. 2017). Alternatively, mosquitoes that attempt to bite a protected

host may abandon the area after brief exposure to volatile pyrethroids and exhibit subsequent irritation or excitability (Achee et al. 2012a). Ultimately, the amount of time the mosquito is actually exposed to pyrethroid volatiles is a crucial factor for understanding behavioral and physiological responses. Reduced time of exposures in an area where volatile pyrethroids are used may be due to multiple factors, such as whether the mosquito spends less time within the toxicant plume before dispersing away (Buhagiar et al. 2017b), or because there is a limit to the speed and extent to which vapors can penetrate into harborage (Buhagiar et al. 2017a).

With this in mind, the extent to which short-duration exposures to volatile pyrethroids can induce sub-lethal effects on mosquito oviposition behavior and reproductive success has not been evaluated. Prior work by our group used 2h of exposure to low concentrations of transfluthrin volatiles to establish a proof-of-concept that volatile pyrethroids have effects on mosquito oviposition behavior and reproductive output (Bibbs et al. 2018a). Prior work under semi-field conditions used shorter exposure durations and did not observed sub-lethal effects when using metofluthrin in domestic environments (Buhagiar et al. 2017b). The lack of sub- lethal effects observed in Buhagiar et al. (2017b) may have been a result of high concentration or inadequate distance of mosquitoes from the point-source to adequately administer a sub-lethal exposure in these studies. These differences among studies provide several questions that need to be addressed before downstream oviposition interference by exposure to pyrethroid volatiles is considered for integrated mosquito management. Alternatively, sub-lethal behavioral or reproductive effects of volatile pyrethroids might just be a useful side effect beyond their repellency when used by consumers as stand-alone devices or operationally in focal treatments.

Therefore, exposing resistant and field strains to spatial repellents is warranted to test the extent to which downstream effects on oviposition behavior and reproductive fitness are similar

to those shown in prior work on susceptible laboratory colonies (Bibbs et al. 2018a).

Additionally, exposure durations need to be restricted to those applicable to real world situations, such as the brief interlude that may occur between a host-seeking mosquito and a spatial repellent protected host. Furthermore, more toxicants besides transfluthrin need to be examined to determine the extent to which volatile pyrethroids collectively have the benefit of causing oviposition or fecundity changes in mosquitoes. To address these gaps in our understanding of sub-lethal effects of volatile pyrethroids on mosquito performance, we evaluated a brief 60 second exposure to metofluthrin vapors at an LC30 concentration, taken from data reported by

Bibbs et al. (2018b), while replicating the experimental design of Bibbs et al. (2018a) to assess effects of exposure on oviposition behavior and reproductive output. With these experiments, we aim to confirm that sub-lethal effects on oviposition behavior and fecundity are observable from exposure to a second volatile pyrethroid, metofluthrin, against multiple mosquito strains that vary in their time in the laboratory and resistance status, and following a short exposure that better represents the brief interaction time of host-seeking mosquitoes and a repellent-protected host.

4.2 Materials and Methods

4.2.1 Insect Rearing

Pyrethroid susceptible, 1952 Orlando, FL strain Ae. aegypti (ORL) and pyrethroid resistant 2012 Puerto Rican strain (PR) were acquired from the United States Department of

Agriculture, Agricultural Research Service, Center for Medical, Agricultural, and Veterinary

Entomology (USDA-ARS-CMAVE) in Gainesville, Florida. A 2017 St. Augustine, FL field strain of Ae. aegypti (STA) was started from eggs collected by the Anastasia Mosquito Control

District of St. Johns County. All colonies were kept at 26 ± 1°C, 85 ± 5% relative humidity

(RH), with a 14:10 (L:D) photoperiod. Batches of 2,000 eggs were placed in larval pans

containing 2,500 ml of reverse osmosis (RO) water. Developing larvae were fed 1-3 g of a liver- yeast powder mixture at a 3:2 ratio ad libitum in a 50-ml suspension. Adult mosquitoes were kept in 30 x 30 x 30 cm flight cages supplied with 10% sucrose solution and RO water.

Mosquitoes used in bioassays were 5-7d old females that were given the opportunity to mate without being blood fed.

4.2.2 Bioassay Design

Technical grade metofluthrin was supplied by McLaughlin Gormley King Company

(96.65% SumiOne, Sumitomo Chemical Company, Ltd., Tokyo). Bioassay protocols were based on methods from Bibbs et al. (2018a; 2018b). In brief, single-use, 473 ml clear polypropylene containers were modified to allow mosquitoes to be aspirated into the chamber. A 5 x 40-mm pleated filter paper was prepared with 40-μl aliquots of either acetone, for negative control, or

0.029% metofluthrin diluted in acetone, for treatment. The 0.029% metofluthrin concentration was selected as a predicted LC30 for ORL using data from Bibbs et al. (2018b), and was the only sub-lethal concentration tested.

Once treated strips had dried for 6-min, 20 females of a given strain were aspirated into an exposure chamber. Treated paper strips were nested in a sachet through the aspiration slot and the chamber sealed for 60s. After 60s of metofluthrin exposure, mosquitoes were transferred into a clean flight cage identical to those used in rearing. Immediately upon transfer, all treatment and control cohorts were offered 48 ml of non-citrated bovine blood according to Siria et al. (2018).

Freshly warmed blood-meals were supplied for 2hr per cohort, with blood replaced each hour.

Fed mosquitoes were allowed 72h post blood-meal to become gravid while held in the rearing conditions. To assess oviposition and fecundity, mosquitoes that could clearly be identified as gravid within each cohort were transferred individually to polypropylene flight cages measuring

30 x 30 x 30 cm for oviposition (Bugdorm I # DP1000, MegaView Science Co. Ltd., Taiwan).

Groups of ten flight cages, each with one mosquito per cage, were used for ORL, STA, and PR strain LC30 treatments, and an additional group of ten cages were used for control females of each respective strain, totaling 60 cages per replicate. All other oviposition bioassay conditions adhered to Bibbs et al. (2018a).

Data collection procedures similarly abided Bibbs et al. (2018a). To summarize, post- bioassay mosquitoes were removed and dissected in 70% EtOH under a light microscope to examine egg retention. Dispersion of eggs laid across cups and the quantity of eggs in each cup were recorded upon removal of adults for dissection to assess egg retention. To facilitate egg embyronation, oviposition papers were stored in rearing conditions and loosely enveloped in wax paper for a 24h damp drying period. Eggs were inspected for any deformities following damp drying and then submerged in 118-ml cups filled with 50 ml of RO water. Larvae were reared post-hatch with larval food provided ad libitum until all larvae pupated or perished. The number of larvae that hatched and subsequent natal mortality were recorded daily over the rearing period.

Unhatched eggs were examined for deformities again once the larval cohort pupated. Eggs were considered non-viable if at either examination the chorion was collapsed. This design was replicated three times for each treatment and control for ORL, STA, and PR.

4.2.3 Data Analysis

Kruskal-Wallis analysis of variance and post hoc Steel-Dwass tests were performed on the number of eggs collected from each treatment cohort, number of oviposition containers with eggs present, number of larvae successfully hatched and reared to 4th instar, number of eggs that collapsed, and the number of eggs retained in dissected adult females after the bioassay. Kaplan-

Meier survival functions were performed on mortality over time for larvae during hatch and rearing. Log-rank tests were used to compare control and treatment mortality curves. Summary

statistics were calculated after correcting values with Abbott’s formula (Abbott 1925). Statistical procedures were performed using JMP 13.1.0 (SAS Institute, Inc., Cary, NC).

4.3 Results

In all strains, untreated Ae. aegypti oviposited in all 6 containers during 90-100% of bioassays. In contrast, all strains exposed to sub-lethal concentrations of metofluthrin vapors deposited eggs across significantly fewer containers than the untreated groups of the

2 corresponding strain (Fig. 4-1, χ (5,120) = 92.7, p < 0.0001). For egg deposition within strains,

2 treated mosquitoes laid significantly fewer eggs than controls (Fig. 4-2, χ (5,120) = 71.2, p <

0.0001), with a 48%, 36%, and 18% reduction in ORL, STA, and PR strains after 60s exposure to metofluthrin respectively. Otherwise, no significant differences in the production of viable

2 eggs were detected between treated ORL, STA, and PR cohorts (χ (5,120) = 10.52, p = 0.0618), indicating no obvious differences in the susceptibility of each strain to the volatile pyrethroid treatment (Fig. 4-2). Once oviposition bioassays were completed, dissections of adults revealed significantly more eggs were retained in the ovaries of treated ORL and STA mosquitoes as compared to controls, but there was no difference between the treated PR strain and its

2 corresponding control (Fig. 4-2, 4-3A, 4-3B, 4-3C, χ (5,120) = 22.9, p < 0.0004). In a singular instance, a full clutch of pre-maturely melanized eggs were dissected out of a treated STA female following the bioassay (Fig. 4-3B).

In addition to egg retention, eggs deposited from treated female mosquitoes across all strains differed in viability. A proportion of eggs from all treated strains were observed to collapse before having the opportunity to hatch (Fig. 4-2, 4C, 4D), while this was never observed

2 to happen in the control groups for all strains (χ (5,120) = 21.6, p < 0.0006). Viability was reduced in the treated ORL, STA, and PR strains by 48%, 24%, and 19% respectively as a result of egg collapse (Fig. 4-2), and although all three treatments significantly differed from their control

2 (χ (5,120) = 19.46, p < 0.0016), none of the pairwise tests between treatment groups were statistically significant after controlling for multiple testing. In ~3% of the STA strain eggs, the chorion failed to completely melanize (Fig. 4-4A). These teneral eggs always collapsed. Beyond egg phenotypes, hatching anomalies (Fig. 4-4B) among treated mosquito progeny resulted in reduced larval survivorship in the F1 generation. The eggs resulting from treated STA, and to a

2 lesser extent treated PR, mosquitoes showed accelerated hatching (Fig. 4-5 χ (5) = 410.3, p <

0.0001), with STA having the greatest occurrence of accelerated hatching (Fig 4-5; p < 0.0019), followed by PR (p < 0.0033), and no observed hatching time anomalies in treated ORL strain or the untreated controls. This accelerated hatch occurred primarily within the first 24h after removing egg papers from the bioassay (Fig 4-4B) while they were held within wax paper envelopes for a drying period, to facilitate embryonation.

Consequently, the associated larval mortality among treatment cohorts when attempting

2 to rear the F1 generation was significantly higher than in the control cohorts (Log-Rank: χ (5) =

1846.6, p < 0.0001), and varied by time from the point of removal from the bioassay until

2 successful rearing to pupation (Fig. 4-5, χ (5) = 14.74, p < 0.0142). After the drying period, all treated cohorts experienced significant reductions in larval survivorship during rearing (Fig 4-5),

2 with STA (Log-Rank: χ (5) = 797.6, p < 0.0001) having greater reduction in survivorship than

2 2 both ORL (Log-Rank: χ (5) = 145.8, p < 0.0001) and PR (Log-Rank: χ (5) = 113.6, p < 0.0001), which were not different from each other. Cumulatively, the corrected larval mortality accounted for a 24%, 60%, and 14% decrease in recruitment to the next adult generation for the treated

ORL, STA, and PR strains, respectively. When reduced egg yield, reduced fertility, and reduced larval survivorship were accounted for, an overall reduction in fitness of 79%, 81%, and 43%

was quantified for the ORL, STA, and PR strains of Ae. aegypti following exposure to metofluthrin volatiles.

4.4 Discussion

Against pyrethroid susceptible ORL strain Ae. aegypti, we confirmed that as little as 60s of exposure to metofluthrin vapors can induce sub-lethal effects that reduce overall fitness through reduced fecundity, fertility, egg dispersion during oviposition, and larval survivorship.

These effects appear stronger than a prior study with transfluthrin, which reported fecundity, fertility, and egg dispersion changes, but not significant larval mortality, after 2h of exposure to vapors (Bibbs et al. 2018a). Additionally, we discovered that wild-type STA and pyrethroid- resistant PR strains phenotypes were not protected from these sub-lethal effects, and had additional larval mortality from premature hatching of the larvae. For context, the PR strain Ae. aegypti has been reported to have fixed V1016I (97% allele frequency) and F1534C (95% allele frequency) kdr sodium channel mutations, in accompaniment with overexpressed cytochrome

P450 and glutathione-S-transferase enzymes, resulting in resistance ratios of 49.9, 112.5, and

127.6 for Type I, Type II, and non-ester pyrethroids, respectively, as compared to the ORL strain

(Estep et al. 2017; 2018). Mosquitoes in the field may be susceptible to changes in fitness and oviposition after surviving exposure to spatial repellent vapors. This is especially encouraging when evaluating the STA field strain, which bears heterozygous pyrethroid resistance traits from the same kdr and cytochrome P450 mutations (Estep et al. 2018) and serves to model a wild-type mosquito.

Although the core outcomes from a previous study with a longer exposure to transfluthrin were conserved when re-evaluating with a short exposure to metofluthrin, there were differences that are worth considering. Metofluthrin caused females of the ORL strain to occasionally perish after consuming a blood meal before having the opportunity to lay eggs. This effect was not

observed in other strains or in previous work with transfluthrin. The extent of ORL strain larval mortality was clearly detectable after 60s of metofluthrin exposure (Fig. 4-5) despite negligible larval mortality when previously treating with 2h of transfluthrin (Bibbs et al. 2018a).

Furthermore, the cause of the early hatching we observed in the STA and PR strains (Fig. 4-4B), with subsequent increases in larval mortality (Fig. 4-5), is unclear as well. One possibility is the concentration required to elicit a neuronal response from pharate larvae within the eggs may be minute compared to the adult mosquito. If the parent female fails to metabolize or sequester the toxicant sufficiently by the time oocytes are prepared to pass through the oviduct, then the eggs may acquire a maternal exposure sufficient to result in signs of toxicity after the larvae embryonate, resulting in larvae that were stimulated to hatch too early.

With the inclusion of these newly reported outcomes, it is clear that metofluthrin has the capacity to induce sub-lethal effects on the fitness of pyrethroid resistant Ae. aegypti strains that are collected from the field (STA) or selected in laboratory (PR). Given the effects we observed with a 60s exposure, it is plausible that sub-lethal effects could be observed in real world scenarios. Previous work indicates that toxicity overshadows sub-lethal effects in larger environments (Buhagiar et al. 2017b), implying that it is unrealistic that sub-lethal effects would be observed in the field. Additionally, the exposure design of our bioassay confines mosquitoes near the point-source and prevents the mosquitoes from avoiding the toxicant upon detection. In semi-field evaluations, mosquitoes that were allowed to freely navigate an air space avoided metofluthrin when the opportunity presented itself. This was observed in both high volume areas where mosquitoes circled around transfluthrin vapors to unprotected areas to reach the host

(Ogoma et al. 2014) and also in tunnel type designs where mosquitoes avoided progressing down a tunnel to reach a host if the tunnel contained metofluthrin vapors (Ponlawat et al. 2016).

Contrasting with the aforementioned examples, the present study was intended to measure the outcomes in the surviving mosquitoes after the low-concentration exposure. However, mosquitoes may detect and avoid metofluthrin at lower concentrations than what was used in these controlled experiments. For example, demonstrations in hut studies used a series of four passive emanators in 12m3 huts, with each emanator covering a 3m3 volume using 3.5ml of a

30% metofluthrin solution per emanator (Stevenson et al 2018). Under the aforementioned conditions, An. gambiae Giles avoided the treated hut and 40% more were collected outside the treated airspace, with general repellency as high as 60% despite the presence of hosts (Stevenson et al 2018). Conversely, when mosquitoes are unable to escape exposure, the toxicity spike in metofluthrin may be such that the mosquito has a greater probability of dying than experiencing sub-lethal effects (Buhagiar et al. 2017b). Therefore, the concentration gradient that results from delivering treatment, the three dimensional physics of the vapors, and the resulting behavior of the mosquito are vital considerations for vapor-based toxicants or repellents when refining their utility.

Despite support on the effectiveness of metofluthrin, we understand there are risks associated with changing the operationally used concentration of toxicants irresponsibly. The

USEPA designated metofluthrin as a suspected as a hepatocarcinogen (EPA 2006) based on rats developing liver tumors following 2-years of chronic exposure to diets of 900 and 1800 ppm

(males) or 1800 ppm (females) of metofluthrin (Deguchi et al. 2009). The tumors were not observed at lower concentrations. The cancer risk assessed by the EPA has been contested because the mode of action for the liver tumors is not present in human physiology, and there was no correlation between metofluthrin exposure and genotoxic effects or downstream cell replication in mammalian tissues (Deguchi et al. 2009, Hirose et al. 2009, Yamada et al. 2009;

2015, Ohara et al. 2017, Okuda et al. 2017). The summative EPA safety data also categorizes the vertebrate LC50 of metofluthrin between 1.08 and 1.96 ppm in air within 24h of exposure (EPA

2006). Additionally, the lowest observed adverse effects in mammals occurred at 0.196 ppm and no observed adverse effects were documented for chronic inhalation exposures of ≤0.099 ppm when exposures were sustained for 4h/d for 28d (EPA 2006). Using the chemical properties reported by the EPA (2006), we can account for the vapor pressure (1.47x10-5 Torr), molecular weight (360.34 g/mol), and concentration of our acetone diluted solution (0.029%), to calculate an expected airborne concentration of 0.3 ppb during our bioassays. This concentration falls well below the documented concentrations for no observed adverse effects. In review of chemical literature, metofluthrin is a type of trans-chrysanthemate pyrethroid (Ujihara et al. 2012) that does not bind effectively to human sodium channels (Smith and Soderlund 1998). This could account for the large therapeutic index between vertebrate and mosquito toxicity, which has been demonstrated to be 10-fold and 100-fold larger than those observed with comparable trans- chrysanthemate pyrethroids, such as prallethrin and d-allethrin respectively (Sugano and

Ishiwatari 2012).

We are confident that there are safe, effective outdoor applications for metofluthrin that can extend beyond the current spatial repellent niche. Outdoor residual treatments and ultra-low volume space sprays are possible modes of delivery that could expand the utility of spatial repellents and could provide tools to large area mosquito abatement programs. Future studies should investigate the concentration gradients emanating from outdoor treatments in order to assess the underlying potential of instigating sub-lethal effects against mosquitoes in the field.

Our study demonstrates that sub-lethal effects can be observed in field-derived and resistant Ae. aegypti strains after 60s of exposure to metofluthrin. With the available safety data, pre-existing

spatial repellent registration, and possibilities for other outdoor delivery methods, metofluthrin is worth developing into field tools for larger mosquito abatement operations.

4.5 Disclosures

This article was published in the Entomological Society of America Journal of Medical

Entomology in 2019, volume 56, pages 1087-1094. McLoughlin Gormley King (MGK)

Company and Sumitomo Chemical Company provided technical grade metofluthrin for this study.

4.6 Figures

Figure 4-1. Reduction in the mean percent usage of oviposition sites by Aedes aegypti (L.) after sub-lethal exposure to metofluthrin vapors. Strains represented are 1952 Orlando strain (ORL), St. Augustine wild type (STA), and pyrethroid resistant Puerto Rican strain (PR), with both treatment (T) and control (C) data. Figures shown with standard error of the mean as I-bars for comparison between treatment (ORL = dotted, STA = diagonal hash, PR= horizontal hash bars) and control (solid bars).

Figure 4-2. Reduction in egg yield and expression of multiple egg phenotypes among three strains of Aedes aegypti (L.) after exposure to sub-lethal concentrations of metofluthrin vapors for 60 seconds. Compound bar graphs represent the counts (n = 30) of collapsed eggs (bottom bar), viable eggs (middle bar), and eggs retained in the parent female (top bar) following 72h of oviposition in a six-cup arena. Strains represented are 1952 Orlando strain (ORL), St. Augustine wild type (STA), and pyrethroid resistant Puerto Rican strain (PR), with both treatment (T) and control (C) data paired for comparison. Figures are shown with 95% confidence intervals depicted as I-bars for comparison between treatment and controls, as well as among different strains. Viable plus Collapsed eggs equate the total eggs oviposited.

Figure 4-3. Multi-plate figure showing sub-lethal effects expressed by Aedes aegypti (L.) 1952 Orlando strain (ORL; plate A), St. Augustine wild type (STA; plate B), and pyrethroid resistant Puerto Rican strain (PR; plate C) following 60s exposure of the parent female mosquito to metofluthrin vapors prior to blood feeding. A: An ORL reproductive tract, following treatment, with 20 or fewer retained eggs. B: Singular event of a treated STA female with retained, melanized eggs following the 72h oviposition bioassay. C: Dissected reproductive tract from a treated PR mosquito that exhibited no egg retention and was indistinguishable from untreated mosquitoes.

Figure 4-4. Multi-plate figure showing egg phenotypes occurring as sub-lethal effects in Aedes aegypti (L.) St. Augustine wild type (STA; plates A and B), Puerto Rican strain (PR; plate C), and 1952 Orlando strain (ORL; plate D) following 60s exposure of the parent female mosquito to metofluthrin vapors prior to blood feeding. A: STA egg cluster containing a teneral egg (arrow) during a 24h damp dry cycle. B: Accelerated hatching of STA with 1st instar larvae (arrows) on ovipaper during a 24h damp dry cycle. Papers were stored in wax sleeves during drying, and no comparative effect was observed in control mosquitoes. C: Cluster of PR eggs from treatment demonstrating variable phenotypes in one clutch (solid circle = viable, dotted circle = collapsed, and dashed circle = accelerated hatch). D: Cluster of collapsed ORL eggs from a treatment oviposition bioassay.

100

Figure 4-5. Kaplan-Meier plots displaying decreases in larval survivorship across the post- bioassay period for treatments and controls of 1952 Orlando (ORL), St. Augustine wild type (STA), and pyrethroid resistant Puerto Rican (PR) strain Aedes aegypti (L.). Day 0 indicates removal of eggs from oviposition bioassays. Day 1-7 was the larval rearing period. Control cohorts are represented by solid gray lines and treatment is represented by hashed red lines. Larval survivorship decreases within the first 24h after removal from the bioassay were representative of accelerated hatching and subsequent natal mortality.

101

CHAPTER 5 OLFACTOMETRIC COMPARISON OF VOLATILE PYRETHROID BARRIER FORMULATIONS USING MOSQUITO BEHAVIORAL RESPONSES

5.1 Introduction

Spatial repellent pyrethroids have a spectrum of toxicological effects, most notably various forms of irritation and repellency (Achee et al. 2012a, Bibbs and Kaufman 2017), including a range of minute concentrations that are lethal (Bibbs et al. 2018b). Active ingredients such as metofluthrin appear to have utility for domestic mosquito control through many delivery methods (Ritchie and Devine 2013, Bibbs and Xue 2015). Unfortunately, metofluthrin and similar volatile pyrethroids are difficult to evaluate and handle consistently (Bibbs and Kaufman

2017), due in part to the low exposure required to kill mosquitoes, sometimes overshadowing attempts to observe behavioral changes (Buhagiar et al. 2017b). Although there is incentive to formulate metofluthrin for area-based management of container inhabiting Aedes mosquitoes, there is no clear evidence of what delivery platforms may serve best for integrated mosquito management. One option is creating a stable formulation that permits an outdoor residual treatment. As a technique, outdoor residual sprays against foliage and harborage are reviewed in a positive light for managing container inhabiting Aedes mosquitoes (Faraji and Unlu 2016).

Formulating metofluthrin as a barrier product provides the basis for transitioning spatial repellents to a role in mosquito abatement programs, but it must be determined how mosquitoes respond to a formulation with strong repellency or strong mortality.

Current methods to test spatial repellents include olfactometry bioassay (Kline et al.

2003, Obermayr et al. 2015), semi-field (Revay et al. 2013, Bibbs and Xue 2015), and field studies (Kawada et al. 2005; 2006; 2008). Olfactometry and wind tunnel bioassays generate information more quickly than field studies, but are conventionally limited to non-toxic volatile compounds (Kline et al. 2003, Obermayr et al. 2015) or compounds with limited volatility

102

(Cohnstaedt et al. 2011). Metofluthrin is toxic to container-inhabiting Aedes exposed to dried deposits following a vapor phase delivery of the active ingredient (Bibbs et al. 2018b). Such sensitivity in mosquitoes complicates the use of acrylics, plexiglass, or polymer plastics that are prone to contamination from pyrethroids, DEET, or kairomone odorants during olfactometry bioassays. Thus, reducing contamination between trials can require intensive cleaning or reconstruction of test chambers to mitigate confounding effects when collecting data. Therefore, there is an apparent need in designing and constructing an apparatus that simplifies the evaluation of toxic spatial repellents. An inert device would enable the exploration of formulated metofluthrin to confirm biological end points that could translate to the field.

The first goal of this study was to evaluate a glass and metal olfactometry apparatus for its suitability to examine mosquito behavior following volatile pyrethroid exposure. The second study goal was to evaluate the extent to which mortality and repellency could be observed when exposing Ae. albopictus to a mix of host volatiles and formulated metofluthrin. The underlying purpose of evaluating formulated metofluthrin was to determine if metofluthrin could successfully prevent the approach of peridomestic mosquitoes to a host upwind of the barrier.

Given the historic view of metofluthrin as a repellent, deterrent (Achee et al. 2012a) and volatile toxicant (Bibbs et al. 2018b), this study was the first step in determining whether metofluthrin when applied as a barrier spray would be efficacious in protecting people from mosquitos bites.

5.2 Materials and Methods

5.2.1 Insect Rearing

Gainesville 1992 strain Aedes albopictus eggs were obtained from the United States

Department of Agriculture, Center for Medical, Agricultural, and Veterinary Entomology in

Gainesville and reared at 26.6°C, 85±5% relative humidity (RH), with a photoperiod of 14:10

(L:D). Oviposition papers holding 2,000 eggs were submerged in larval pans filled with 2,500ml

103

of reverse osmosis (RO) water. Hatched larvae were fed 1-3g of liver and yeast mixture at a 3:2 ratio ad libitum in a 50-ml suspension. Emerged adults were housed in flight cages and provided with supplies of 10% sucrose solution and RO water. All mosquitoes used in this study were 5–

7d old, non-blood-fed, female Ae. albopictus.

5.2.2 Modular Wind Tunnel

Trials were conducted in a glass and metal, dual-port, modular wind tunnel (MWT) (Fig.

1) to which the overview of the prototypical model are reported here; iterations of the MWT are a product of Sigma Scientific, LLC with detailed descriptions available upon request (Estaver and Strohschein 2019). The flight tunnel consisted of structural aluminum support beams (30-

3030, 80/20 Inc., Columbia City, IN) and soda-lime glass (Table glass, Shea’s Glass Co.,

Gainesville, FL) sides, top, and bottom panels. This module formed the center section at 50mm tall by 520mm wide by 1560mm long and was connected to front and rear modules of equivalent cross-section dimensions (500mm x 520mm). The panels were rigidly fixed in the frame, so removal of the front and rear modules was required for cleaning. A 260mm long specimen release module was attached downwind of the flight tunnel module. Once latched on to the flight tunnel, the specimen release module (Fig. 1, left) allowed air to flow over the subjects and kept them from progressing up the tunnel until experiment initiation. The module terminated in a screened aluminum end to allow for the air to exhaust out of the MWT. The 520mm long plenum module connected the back of the MWT to 10cm diameter ductwork that exhausted the stimuli- laden air from inside the tunnel to the outside of the test space (Fig. 1, left). This ensured that contaminated air was not recycled through the MWT. To provide consistent airflow, a blower assembly (DC OEM Specialty Blower 3HMH7, W. W. Grainger, Inc., Lake Forest, IL) was connected to the exhaust ductwork and mounted outside of the laboratory. Laminar flow with no eddy current within the flight tunnel was achieved by the blower enabling a slight negative

104

pressure within the MWT. The blower assembly was regulated using a variable speed DC motor

(90-240 Volt 50/60Hz power, W. W. Grainger, Inc., Lake Forest, IL) and a pulse width modulator that allows a range of 0.05m/s – 0.3m/s airflow. For standard operation, the wind tunnel was defaulted to 0.2m/s air flow to emulate other wind tunnels and olfactometers.

Upwind of the flight tunnel, the 470mm long bifurcation module contained bulkhead fittings for 0.64cm aluminum tubing (Fig. 1, right). The front of the module had a perforated plate (43% open area) and screening to prevent mosquitoes from traveling further up the MWT, forcing mosquitoes to choose between two ports in the module. Ports contained in the perforated plate contained two glass tubes (150mm outer diameter), each of which had an inverted metal screen cone on the end facing the flight tunnel and an indented closed screen on the opposite end.

At the center rear of each glass tube was the 0.64cm metal tubing that released the stimuli into the air stream. The flow rate of the stimuli was set by flowmeters to match the internal air velocity. Air drawn through the module using the externally mounted exhaust fan (Fig. 1, left) was forced to pass through a filter module at the opposite end of the MWT from the blower assembly (Fig. 1, right).

The filter module latched on to the end of the system and contained perforated aluminum plate and screening as a secondary barrier to prevent mosquitoes from reaching the filters. The module then expanded to a 650mm wide by 575mm tall filter holder. Two activated carbon filters measuring 56cm by 64cm with 2.5cm pleats were placed in separate holders. Attached to the activated carbon filter holder was a 10cm deep, pleated MERV 13 filter with 957cm2 of surface area (FC100A1037s, Honeywell International, Inc., Charlotte, NC). The combined action of the 20cm depth of filter material and the subsequent aluminum screening and perforated plate work together to laminarize the flow of air as it is drawn through the filter module and mitigate

105

air turbulence to be below detectable levels with a digital anemometer. Lack of turbulence was confirmed with a smoke and laser test, returning a “dead calm” reading based on the Beaufort scale, where smoke rises vertically despite the turbulence.

Stimuli were introduced with an externally attached glass chamber with two glass treads for attaching 0.64cm diameter polypropylene tubing, one for incoming air and one directly connected to the bulkhead on the bifurcation module. The stimuli chamber was made of borosilicate glass and had an iris port on one end to allow placement of baited lure or other odor sources without contact with test subjects (Fig. 1, top). Odorants were fed into the existing airflow of the wind tunnel using a clean air delivery system (CADS 2-push & ICAF 2x6, Sigma

Scientific LLC, Micanopy, FL) that purified incoming air of particulates, volatile organic compounds, and pollutants such as SOX, NOX, CS2, etc.

5.2.3 Procedural Validation

To give confidence to procedures used in bioassay, smaller technical investigations were made to seek out confounding effects in the MWT. Metofluthrin formulated for outdoor residual insecticide treatment was supplied by McLaughlin Gormley King Company (32% a. i. emulsifiable concentrate: EXP141610001, Sumitomo Chemical Company, Ltd., Tokyo, Japan).

The MWT was warmed to 26±1°C, humidified to 65±5% RH, and calibrated to dispense filtered air at a flow rate of 2L/min. For the first validation test, toxicant removal by UV light was assessed by applying a 20ml aliquot of undiluted metofluthrin formulation (32% metofluthrin,

EXP141610001) to 10×10cm (100cm2) muslin cloth. The cloth was air dried in a fume hood for

15 min, until dry to the touch, and then nested in a headspace collection chamber with air from the treatment passing through both ports on the wind tunnel for 20min. After contaminating the

MWT, airflow was shut off and 25 host-seeking, female Ae. albopictus were placed in the MWT for a 20-min acclimation period while contained within a mesh topped 500ml polypropylene cup.

106

The cup was handled using gloves to avoid depositing skin odorants on the container. After acclimation, the airflow was re-connected to both ports and administered with a 0.2m/s air velocity through the bulkhead until odorants were taken up by the 2L/min airflow of the MWT.

At this time, a wire was used to pull the screen off of the mosquito container without needing to reopen the wind tunnel to outside air. Mosquitoes were exposed to the contaminated wind tunnel for 20-min before being recollected with a vacuum aspirator and held in a flight cage with access to 10% sucrose solution for mortality assessment at 24hr. For comparison, the MWT was then exposed to a 15-min UVC light cycle using a 4-ballast assembly with high powered bulbs

(FR40T12G4, Solacure LLC, Browns Summit, NC). A second cohort of 25 mosquitoes was acclimated with the same specifications the contaminated replicate. After acclimation, mosquitoes were allowed 20min to fly in the MWT before they were recollected by vacuum aspirator and stored separately from other mosquitoes, also with access to 10% sucrose solution.

Mortality was assessed after 24hr, as with the toxicant exposed cohort. Tests were repeated 3 times.

The second validation involved delivering attractant laden air through both ports. For this test, the MWT was again warmed to 26±1°C, humidified to 65±5% RH, and calibrated to dispense filtered air at a flow rate of 2L/min. Filtered air was mixed with headspace collections from a chamber that contained skin lure (BG-Lure™ cartridge, BioGents AG, Regensburg,

Germany) and 100g of CO2 sublimating from dry ice to create a host attraction mixture (Fig. 1).

The MWT was prepared with the same specifications as the prior validation while excluding any toxicants. Cohorts of 25 mosquitoes were released from acclimation and allowed 20min to choose between either port. The test was repeated after the wind tunnel was rotated 90° and 180° with all other aspects repeated. Each orientation was replicated 3 times.

107

5.2.4 Behavioral Assay

Metofluthrin formulated for outdoor residual insecticide treatment was supplied by

McLaughlin Gormley King Company (32% emulsifiable concentrate: EXP141610001,

Sumitomo Chemical Company, Ltd., Tokyo, Japan). Onslaught Fast Cap (6.4% esfenvalerate,

1.6% prallethrin, 8% PBO Onslaught Fast Cap, McLaughlin Gormley King Company,

Minneapolis, MN), hereafter referred to as ‘Onslaught,’ was selected as a comparison treatment because it was a residual insecticide formulated with prallethrin, a commercially available volatile pyrethroid currently used in mosquito adulticides. Olfactometry was performed in the

MWT to discriminate attraction to host lures, repellency by treatments, knockdown, and mortality of exposed mosquitoes. As in the prior tests, the MWT was warmed to 26±1°C, humidified to 65±5% RH, and calibrated to dispense filtered air at a flow rate of 2L/min. Filtered air was mixed with headspace collections from a chamber that contained skin lure (BG-Lure™ cartridge, BioGents AG, Regensburg, Germany) and 100g of CO2 sublimating from dry ice to create a host attraction mixture (Fig. 1). The non-toxic treatment was the attractant mixture by itself. For toxic treatments, metofluthrin (32% EXP141610001, Sumitomo Chemical Company,

Ltd., Tokyo, Japan) was diluted at 8.7ml/L (1oz/gal) of water and a 20ml aliquot of the diluted treatment was applied to 10×10cm (100cm2) muslin cloth. The cloth was air dried in a fume hood for 15 min, until dry to the touch, and then nested in a second inline headspace collection chamber (OSI, Sigma Scientific, LLC, Micanopy, FL) connected to the attractant blend farther downwind of the supply line (Fig. 1). Comparison was made with Onslaught Fast Cap (6.4% esfenvalerate, 1.6% prallethrin, 8% PBO Onslaught Fast Cap, McLaughlin Gormley King

Company, Minneapolis, MN), which also was diluted at 8.7ml/L (1oz/gal) of water, with a 20ml aliquot of the diluted treatment applied to a separate cloth. The attractant-primed air was delivered into one olfactometer port, leaving the second port with no attractant or chemical

108

treated air to specifically target whether treatments interfere with upwind approach to host lures, rather than driving mosquitoes into an alternative choice. Mosquitoes entering each port were prevented from reentering the wind tunnel because of the attached conical screen. A second screen capped the terminal end of the port to prevent passage upwind into the filter module.

Each replication was performed on a different day and consisted of a multistep exposure and scoring procedure. Prior to beginning a trial, 60 host-seeking, female Ae. albopictus were placed in the MWT and allowed a 20-min acclimation period using a 500ml polypropylene holding cup covered in a mesh screen. The cup was handled using gloves. After acclimation, the attractant blended airflow was connected to the port and administered with a 0.2m/s air velocity through the bulkhead for one port, decided at random for each replicate, until odorants were taken up by the 2L/min airflow of the entire MWT. At this time, a wire was used to pull the screen off of the mosquito container without needing to reopen the wind tunnel to outside air.

Tests were conducted for 20-min with mosquito responses measured through behavioral scoring outcomes: attracted, repelled, disoriented, and knocked down. Within each of these groups, mosquito mortality was assessed after 24h. Specific to each scoring category, mosquitoes that entered the port cage were considered attracted; mosquitoes that did not enter the port in the presence of a treatment and travelled to the opposite end of the MWT, clinging to the screen separating the exhaust from the specimen release module, were categorized as repelled; mosquitoes that entered the port with no airflow were categorized as disoriented; mosquitoes that were immobilized while still living were scored as knocked down. Remaining mosquitoes that were still active in the wind tunnel but not clearly scoreable based on the aforementioned criteria were not attributed to any category. Scoring occurred immediately concluding the 20-min bioassay, after which mosquitoes were vacuum aspirated out of the olfactometer and transferred

109

to separate flight cages containing a 10% sucrose solution Aspiration was targeted by scoring category based on location within the wind tunnel. Attracted mosquitoes were trapped in the active olfactometer port. Disoriented mosquitoes were trapped in the inactive olfactometer port.

Knocked down mosquitoes were prone on the bottom of the wind tunnel. Repelled mosquitoes were clinging to the exhaust panel opposite the rest of the wind tunnel. Mosquitoes were vacuum aspirated directly from these areas following scoring and checked for mortality after 24hrs within their respective groups. Between replicates, the MWT was purged of chemical residues by running a 15-min UVC light cycle on the interior surfaces of the entire apparatus using 4 high powered bulbs (FR40T12G4, Solacure LLC, Browns Summit, NC). This bioassay and scoring process was repeated 15 times (n= 60; 15 replicates) for each of the treatments.

5.2.5 Data Analysis

Procedural validations were analyzed using student’s t-test. For the UV cleaning validation, mean mortality of mosquitoes between uncleaned and UV cleaned assays were compared. The second validation compared mosquitoes between the two ports for each of the 0°,

90°, and 180° orientations. For behavioral bioassays, analysis was corrected for mortality in the non-toxic attractant blend (Abbot 1925). Analysis of variance was conducted for the 20-min olfactometry trials using Ae. albopictus, whereby data were blocked by treatment preparation, recognizing treatment groups [(experimental treatments: metofluthrin, Onslaught) and attractant] as the main effects and the averaged percent arcsine response [square root (attraction, repellency, disorientation, or knockdown)] as the dependent variables. The average percent mosquito mortality 24hr after concluding the olfactometer trials was analyzed using the same blocking approach with treatment groups [(experimental treatments: metofluthrin, Onslaught) and attractant] as the main effects, but with the averaged arcsine mortality [square root (proportion of

110

mortality)] as the dependent variable. Means were compared using Tukey multiple comparison tests. Analyses were conducted using JMP 13.1.0 (SAS Institute, Inc., Cary, NC).

5.3 Results

The glass and metal modular wind tunnel provided an effective means for discrimination of attraction and repellency for Ae. albopictus (Fig. 5-2). The UV cleaning methods between treatments were effective for reducing contamination from metofluthrin, with an average of 78% mortality among Ae. albopictus when the MWT was not UV cleaned after using a toxicant, as compared to 4% after UV cleaning (Fig. 5-3, t = 9.21, p < 0.0116). The main caveat is that the pliable tubing that extends from the headspace collection and any additional in-line headspace attachments (Fig. 5-2) still need to be replaced after every trial where contaminants, such as with toxicants, are unacceptable. Such a replacement is minimal relative to the cleaning results with the remaining components. In the original efforts, it was postulated that a combination of ozone gas and UV light would be necessary to reduce spatial repellent contaminants below a biologically relevant threshold. In practice, a 15-20min UV light cycle with bulbs nested inside the modules (similar in principal to a biosafety cabinet, Fig. 5-3) was sufficient to remove spatial repellents to an extent that no longer killed or appeared to agitate mosquitoes that were secondarily introduced into the unit. The decontamination bulbs utilized a partial emission into the UVC spectra (Fig. A-1), which may account for the lack of additional decomposition stress required to purge contaminants. The UV cleaning cycle, or if desired a UV and ozone cycle for accelerated cleaning, would be more efficient and practical than manual cleaning with solvents and washing, but all methods are tenable with the current design. Bias did not appear between ports at any orientation when attempting to observe positional effects (Fig. 5-3).

In behavioral bioassays, the five behavioral responses recorded from 20-min olfactometry are represented in Fig. 5-4. Attraction of Ae. albopictus significantly differed across treatments,

111

with the host lure attracting 90% of released mosquitoes, Onslaught blended with host lures attracting 75% of mosquitoes, and the metofluthrin blended with host lures attracting less than

10% of mosquitoes (Fig. 5-4; F2,44 = 561.61, p < 0.0001). Meanwhile, the opposite trend was observed with repellency with none observed to the host attraction lure without toxicants.

Repellency differed between treatments (Fig. 5-4) (F2,44 = 214.74, p < 0.0001) with 10% to

Onslaught plus lure and 62% to metofluthrin plus lure. Additionally, there were differences between treatments in terms of disorientation (F2,44 = 60.64, p < 0.0001) with a high of 12% in the presence of metofluthrin. Knockdown also differed between treatments (F2,44 = 130.55, p <

0.0001) with a high of 18% in metofluthrin treatments. When evaluating post-bioassay survival, less than 1% mortality was observed among the mosquitoes attracted to the lure alone.

Treatments containing Onslaught had mortality ranging between 3-9% among repelled and attracted mosquitoes. In contrast, mortality of metofluthrin-exposed mosquitoes differed between the behavioral outcomes (F2,44 = 40.66, p < 0.0001) with 50%, 37%, 25%, and 95% mortality observed for attracted, repelled, disoriented, and knocked down mosquitoes, respectively.

In general observations of the mosquitoes during trials, upwind flight typically occurred within the first 10min in presence of the host odor lures alone. When trapped in the treatment port, frequent probing through the mesh to reach the bulkhead and odorant tubing was observed.

For both Onslaught and metofluthrin trials, mosquitoes needed the full 20-min trial period if they were to reach the bifurcation ports. Mosquitoes trapped in the treatment port were agitated with frequent flying during both Onslaught and metofluthrin trials. Despite that both ports remained open during all trials, regardless of treatment, it was observed that mosquitoes would only refuge in the untreated port during the metofluthrin trials. Mosquitoes trapped in the untreated port and scored as disoriented were sedentary, groomed often, and did not tend to fly around the interior

112

of the port. Although mosquitoes were scored as disoriented, the behavior could also be indicative of repellency that forced mosquitoes into the untreated port after orienting upwind towards the host cues. Mosquitoes that were repelled to the opposite end of the flight tunnel by either Onslaught or metofluthrin were found to cling to the perforated metal separating the exhaust fan from the specimen release module without scattering or attempting to reorient upwind to the lure. In all trials of each treatment type, some mosquitoes remained within the flight tunnel and made no choice. Trials exposing mosquitoes to metofluthrin resulted in the mosquitoes that remained in the body of the flight tunnel being knocked down. Knockdown was never observed in other treatments, even if agitation was apparent.

5.4 Discussion

Using the current wind tunnel design for this study, it was documented that both repellency and mortality, along with general deterrence through disorientation and knockdown, was observed with the experimental formulated metofluthrin residual spray. The repellency and excitation of mosquitoes in response to the Onslaught formulation was attributed to the 1.6% prallethrin in the mixture, which appears to provoke some avoidance of the residual and the associated vapors (Abdel-Mohdy et al. 2008). The repellency caused by exposure to the

Onslaught formulation was vastly outperformed by the repellency caused by exposure to the metofluthrin formulation, with prior investigations supporting a stronger volatility of metofluthrin than prallethrin (Bibbs et al. 2018b). These findings agree with behavioral analysis targeting prallethrin, whereby the lack of volatility in prallethrin confounded the expectation that mosquitoes would become agitated and have increased exposure to adulticides during flight

(Dye-Braumuller et al. 2017). The magnitude of difference in volatility observed with the metofluthrin formulation and the Onslaught formulation may indicate that metofluthrin could be

113

a useful secondary active ingredient, similar to prallethrin in products such as Duet

(Sumithrin/Prallethrin) and Onslaught.

There were no distinct trends in the proportion of mortality observed within each scoring category from metofluthrin trials, however over 40% of the total mosquitoes scored perished by the 24hr evaluation, regardless of which behavioral outcome was scored. All four scoring groups had mortality, including the repelled mosquitoes, which supports the merit of metofluthrin as a self-inclusive push-kill toxicant (Achee et al. 2012a, Bibbs and Kaufman 2017, Bibbs et al.

2018b). The highest overall proportion of mortality was observed in knockdown, followed by attracted, repelled, and then disoriented mosquitoes. The behavioral similarity of the mosquitoes whose orientation to the lure were disrupted, leading them to enter the untreated olfactometer port, relative to prior descriptions of confusion in literature prompted scoring the subjects as disoriented. Disorientation or confusion is a poorly understood occurrence, as mosquitoes apparently lose track of the host, despite proximity, without clear explanation for the behavioral change (Ritchie and Devine 2013, Buhagiar et al. 2017a). It was not part of the objective to explore disorientation in this study, but the occurrence was notable and should be studied further.

It is arguable that designing bioassays as a noncompetitive presentation of a single stimulus blend can result in abnormal behavior, particularly that behavioral effects may be amplified as compared to what would be observed in the field (Obermayr et al. 2015). However, the novel metofluthrin formulation used in this study was intended to intercept mosquitoes as they approached the host amidst the co-mingled stimuli of attractants and toxicants; i.e. an outdoor barrier application. Therefore, this early investigation appears to favor that residual formulations may interfere with upwind host seeking. The observations made during olfactometry bioassays are a preliminary investigation, as mosquitoes in the wild would be

114

expected to differentiate complex odorant blends in the environment (Nikbakhtzadeh et al. 2014,

Dekel et al. 2016). In consideration of area-wide management, strong repellency in the bioassay results for metofluthrin could indicate push effects in the field.

Regardless of repellency, the fact that a range of biological responses occurred in concert as a result of exposure to the residual formulated metofluthrin is worth studying in the field. Any field operation evaluating residual metofluthrin efficacy needs to account for push effects during surveillance. Repellent compounds, such as metofluthrin, have been suspected of inadvertently pushing mosquitoes to unprotected individuals and failing to resolve the overarching issue of stopping vector contact (Moore et al. 2007). Measuring the outcome of push effects is a rigorous process nuanced with the interactions of natural stimuli and the treatments being used (Cook et al. 2007). Furthermore, tracking other effects, particularly disorientation, in the field will require specific data collection schemes, as the effect is poorly understood (Ritchie and Devine 2013,

Buhagiar et al. 2017a). Nonetheless, the design of inert olfactometry methods, such as the modular wind tunnel designed for these bioassays, is a stepping-stone for better evaluation of spatial repellents. The ability to add an additional tool to the screening process of volatile compounds will help reduce future trial and error when discerning effective spatial ingredients.

5.5 Disclosures

This article was published in the Entomological Society of America Journal of Medical

Entomology in 2019, volume 56, issue 6 in print to date. Funding for this research was provided in part by the Florida Department of Agriculture and Consumer Services: Florida Coordinating

Council on Mosquito Control research subcommittee projects 024377 and 025365. McLoughlin

Gormley King (MGK) Company and Sumitomo Chemical Company provided formulated metofluthrin and Onslaught Fast Cap for this study. Sigma Scientific, LLC sourced the material and labor for constructing the modular wind tunnel.

115

5.6 Figures

Figure 5-1. Diagramatic breakdown of a modular wind tunnel, designed of glass and metal, intended for olfactometry using volatile pyrethroids against mosquitoes. Attractant mixtures were assembled into a headspace collection jar and the stimuli laden air fed to a second chamber. The downwind treatment container housed vapors of a pyrethroid product and allowed mixing before the combined air is injected into one port of the bifurcation module. Negative pressure is created in the flight tunnel by using the plenum module to exhaust air and create a slight negative pressure. Air pulled into the flight tunnel from upwind was filtered from outside contaminants with the filter module, while mosquitoes were acclimated in the specimen release module. Once acclimated, mosquitoes were allowed to fly upwind of the flight tunnel and choose whether to approach the attractant/treatment mixture, flee downwind, or refuge in the second port that contains only ambient air.

116

Figure 5-2. Modular wind tunnel design that prevents chemical memory build-up across bioassays. The realized modular wind tunnel was constructed of soda-lime glass and structural aluminum, including metal gaskets in the interior faces of the modules. Note that the white plastic line extending from treatment mixtures (partially indicated by arrows for attraction mixture) were replaced between toxicant trials to eliminate chemical memory possibilities. All other exposed surfaces in the unit interior could be reached easily with hand washing, ozone gas generators, or UV light cleaning methods.

117

Figure 5-3. Procedural validation with a modular wind tunnel. Top: Mean mortality among adult, female Aedes albopictus (Skuse) after 20-min exposures to metofluthrin contaminants in a modular wind tunnel (MWT). Comparison group is mosquito mortality after 4 high powered UVC bulbs were used for 15 minutes (top right) to purge contaminants. Bottom: Mean Ae. albopictus response to skin lure and CO2 emitted from both ports on the MWT (bottom right) at the default orientation (0°) or rotated 90° and 180° to investigate possible bias.

118

Figure 5-4. Mean behavioral response of Aedes albopictus (Skuse) after olfactometry bioassays using skin lure and CO2 blended with purified air (“Host Lure”), or a 20ml aliquot of either 6.4% esfenvalerate/ 1.6% prallethrin/ 8% PBO after diluting at 8.7ml/L of water (“Lure + Onslaught Fast Cap”) or 32% metofluthrin (EXP141610001, Sumitomo Chemical Company, Ltd., Tokyo, Japan ). Mosquitoes trapped in the port releasing the blend were attracted. Mosquitoes retreating to the opposite end of the olfactometer were repelled. Mosquitoes entering into the second port that lacked any lure or treatment were considered disoriented. Mosquitoes that were immobilized while still living were knocked down. Total bar height represents the mean response (%) among the total number of released mosquitoes (n=60, 15 replicates) by category. Black sections of the bar represent the mean percentage of mortality within each category after 24hr.

119

CHAPTER 6 EVALUATION OF VAPOR-ACTIVE PYRETHROIDS AS AN OUTDOOR RESIDUAL TREATMENT (BARRIER) AGAINST MOSQUITOES

6.1 Introduction

Volatile pyrethroids are a group of compounds that have thus far been categorized as spatial repellents as they prevent vectors entering into an area that envelopes one or more people

(Achee et al. 2012a). Spatial repellents are primarily delivered through personal devices (Dame et al. 2014), which constrain their utility when attempting larger scale operations. For example, a study on metofluthrin mosquito coils required every treated residence of an affected village to receive and use a specific device to functionally protect the residents from malaria (Syafruddin et al. 2014). Such logistical hardships highlight that delivery systems for volatile pyrethroids need to be improved so that these chemicals can function in area-wide operations. Regardless of the drawbacks, volatile pyrethroids have been advocated for urban vector management because they cause multiple damaging effects in mosquitoes (Ritchie and Devine 2013, Bibbs and Xue 2015,

Bibbs et al. 2018a).

In the aforementioned malaria study, malaria levels were reduced by 52% following spatial repellent coil use, but the reduction in mosquito biting pressure was 32% (Syafruddin et al. 2014). The disproportionate rate of malaria transmission was attributed to fewer actively- infected plasmodium vectors post treatment (Syafruddin et al. 2014), and could result from vector mortality after spatial repellent exposure. In studies evaluating other spatial repellent devices, it was reported that volatile pyrethroids caused significant mortality against colonized

Aedes albopictus (Skuse), Anopheles albimanus Wiedemann, and Culex quinquefasciatus Say

(Xue et al. 2012a, Bibbs and Xue 2015), and wild caught Psorophora ferox Humboldt, Ps. columbiae Dyar and Knab, Ae. atlanticus Dyar and Knab, and Ae. taeniorhynchus Wiedemann

(Bibbs et al. 2015). Building on this, Bibbs et al. (2018b) examined vapor dose responses of the

120

active ingredients against multiple mosquito vectors using no-contact assays and found that volatile pyrethroids containing polyfluorinated alcohols, namely candidates such as metofluthrin and transfluthrin, caused 90% mortality at less than 0.255 g/100 ml when tested against Ae. aegypti, Ae. albopictus, Cx. quinquefasciatus, and An. quadrimaculatus Say. Data suggest that in addition to being spatial repellents, some volatile pyrethroids, such as transfluthrin and metofluthrin, are more toxic to mosquitoes than currently marketed alternatives, such as prallethrin (Bibbs et al. 2018b).

The discussed studies demonstrate that volatile pyrethroids are potent toxicants that may influence epidemiological outcomes if they can be used strategically. In unrelated work, prallethrin was supported as an effective ultra-low volume (ULV) adulticide during vector abatement operations in urban areas against Ae. albopictus (Farajollahi et al. 2012, Unlu et al.

2014), providing evidence that other delivery methods are effective with volatile pyrethroids.

However, the lethal concentration to reach 90% mortality (LC90) of polyfluorinated compound vapors were 20-fold more toxic than prallethrin vapors against Ae. albopictus and other tested species (Bibbs et al. 2018b). Furthermore, volatile chemistries such as metofluthrin could maintain resilience against resistant phenotypes. For example, pyrethroid resistant and wild type

Ae. aegypti strains were shown to be as susceptible as laboratory strains to sub-lethal fecundity reduction and oviposition behavioral shift, in addition to heightened larval mortality in the F1 generation following exposure (Bibbs et al. 2019a). These findings indicate metofluthrin could be used in other aduticiding strategies. One option would be as an external residual spray, also called a barrier treatment, whereby the mosquito is fatally separated from an otherwise vulnerable population of hosts through vegetation or structures treated with long-lasting residuals

(Cilek 2008).

121

To be adopted by mosquito abatement programs, barrier treatments would need to maintain sufficient integrity over several weeks to provide sustained control of local mosquitoes while requiring no additional chemical inputs (Cilek 2008). Investigations of bifenthrin, lambdacyhalothrin, and deltamethrin as residual application treatments in semi-field trials have resulted in five weeks of 75-83% reduction in Cx. quinquefasciatus and Ae. albopictus (Cilek 2008, Cilek and Hallmon 2008). In field operations, a bifenthrin barrier applied as the only adulticide in the perimeter of an 11-ha cemetery resulted in 97%, 65%, 58%, and 43% reduction in Ae. albopictus adults with similar reductions in eggs over four successive weeks, respectively (Bibbs et al.

2016). Staggered applications of prallethrin thermal fog treatments and bifenthrin barrier sprays in the vegetation of a 5-ha residential site resulted in three weeks of 100% reduction in Ae. albopictus adults (Gibson et al. 2016). However, despite the success of other synthetic pyrethroids in residual sprays and the inclusion of a volatile pyrethroid in ULV and thermal fogging strategies (Farajollahi et al. 2012, Gibson et al. 2016), volatile pyrethroids have not been tested in barrier applications. Environmentally sensitive active ingredients that remain effective in low concentrations or that can hold-fast against sun and rain exposures are ideal qualities for such long-lasting residual treatments (Cilek 2008, Cilek and Hallmon 2008).

Despite supporting work that indicates metofluthrin is highly toxic to multiple mosquito genera (Bibbs et al. 2018b), it is used in the United States and other areas of the world only as a spatial repellent. Behavioral analysis using olfactometry with a long-lasting metofluthrin formulation was conducted against Ae. albopictus and confirmed a suite of beneficial effects, including lowered attraction to host cues, repellency, disorientation, knockdown, and subsequent mortality (Bibbs et al. 2019b). Although prior findings are positive, long-lasting metofluthrin needs to be deployed as a barrier treatment in field environments and the longevity needs to be

122

tested. Additionally, the relative contributions of repellency versus mortality needs to be confirmed in the field to acertain whether there is a risk of pushing mosquitoes to unprotected areas.

6.2 Materials and Methods

6.2.1 Insect Rearing

Gainesville 1992 strain Aedes albopictus eggs were obtained from the United States

Department of Agriculture, Center for Medical, Agricultural, and Veterinary Entomology in

Gainesville and reared in insectaries maintained at 26.6°C, 85±5% relative humidity (RH), with a photoperiod of 14:10 (L:D). Oviposition papers holding 2,000 eggs were submerged in larval pans filled with 2,500ml of reverse osmosis (RO) water. Hatched larvae were fed 1-3g of liver and yeast mixture at a 3:2 ratio ad libitum in a 50-ml suspension. Emerged adults were housed in flight cages and provided with separate supplies of 10% sucrose solution and RO water. All mosquitoes used in this study were 5–7d old, non-blood-fed, female Ae. albopictus.

6.2.2 Aging Bioassays

Aging bioassays were conducted in the laboratory to determine the anticipated longevity of a 32% metofluthrin formulation (EXP141610001, McLaughlin Gormley King Company,

Minneapolis, MN). Substrates for treatment were 25×25cm (625cm2) sections of cut wood planks. Six wood panels were assigned to each of eight environmental ageing conditions and a control environment, totaling 54 substrates. Formulated metofluthrin was diluted at a rate of

8.72ml of product per liter of water to make a 0.25% solution and applied at 12.4ml per 100cm2, as per label directions. Applications were made using a battery powered backpack sprayer (REC

15 ABZ, Birchmeier Sprühtechnik AG, Stetten, Switzerland) set to a 600kPa flow pressure and fitted with the default cone nozzle. Fifty-four substrates were treated individually with 20ml of treatment and air-dried for 15min. The six substrates assigned to control were treated with RO

123

water and no metofluthrin. All substrates were then housed in the assigned holding conditions and measured repeatedly for the duration of the study. Holding conditions were assigned the following labels: Control, no treatment or special conditions; ‘L’ represented 24hr lighting, ‘D’ represented 0hr lighting, ‘L/D’ represented 12hr lighting; subscripts of <30 represented under

30% humidity and >80 represented over 80% humidity; ‘R2cm’ represented 2cm of simulated weekly rainfall; ‘Outdoor’ represented unregulated exposure to ambient outdoor temperature, humidity, and rainfall.

Substrates were subjected to various conditions during the study to evaluate product degradation among the treatments (Table 6-1, Fig. 6-1). Environmental chambers were constructed using 35.6×27.9×5.9cm clear polyethylene containers (#1965, Sterilite Corporation,

Townsend, MA). For chambers designated for 80% or greater humidity, containers were filled to a 5cm depth of perlite and 3cm depth of water dispersed into the perlite layer. Chambers having

30% or less humidity were filled to a 5cm depth with silica desiccant. A temperature/hygrometer gauge (AX-AY-ABHI-100322, ASX Design, Chicago, IL) was affixed to the inside of each environmental chamber to monitor the stability of humidity. Additionally, an aluminum pan was used to separate the treated substrates from the perlite and silica desiccant while in containment.

For light and temperature maintenance, chambers were stored at 26±1°C ambient temperature in separate incubators (818 Illuminated Low Temperature, Precision Scientific Inc., Winchester,

VA) according to three ratios of light hours to dark hours (L:D) (Table 6-1). For chambers requiring simulated rainfall, treated substrates were removed from their environmental chambers every 7d and doused with water. Rainfall was simulated using a clean battery powered backpack sprayer with the same specifications as the equipment used for treatment. Water was delivered with even passes in 30-sec bursts alternating with 30-sec rests until the reservoir emptied. The

124

last series of substrates were kept outdoors in full exposure to natural heat and rainfall in St.

Augustine, FL from the months of August to October.

To assess the effectiveness of the aged metofluthrin treated substrates, assays were conducted using laboratory-reared mosquitoes on all substrates from each holding condition and the control. Test cages were assembled from 2.5×30cm lengths of cardstock connected at distal ends to form a ring with interior openings. A 10-mm opening was created in the cardstock to serve as an aspiration slot when transferring mosquitoes. The ring was then enclosed on the two interior, open faces using fitted cuts of polypropylene mesh. Ten adult female Ae. albopictus were aspirated into each cage, followed by sealing the aspiration slot with clear tape to prevent escape. Assays were conducted for 20min by setting three test cages with a mesh side resting against the treated face of each substrate, totaling 18 cages per holding condition and each substrate serving as a replicate (n = 30; 6 replicates) (Fig. 6-1). Test cages were removed from substrates after 20min and mosquito knockdown was scored. Exposed cages were then removed from testing and held under aforementioned rearing conditions until was mortality scored at 24h post exposure. Assays were repeated once every week until mosquito mortality no longer exceeded an average of 80% for two consecutive weeks.

6.2.3 Field Study Area

Selected study sites were 0.8ha areas that measured approximately 120×60m at the perimeter and contained residential parcels. The midline of the site was composed of dense vegetative harborage suitable for application of a barrier treatment (Fig. 6-2). The study sites also generally contained abundant mosquito harborage for a variety of mosquito species. Twenty four sites, 18 for treatment replications and 6 for control replications, were selected with these parameters. All treatment sites were separated by a buffer of 3km from an associated control site, with treatment sites separated by more than 3km from other treatment sites.

125

6.2.4 Surveillance

Adult mosquito surveillance was conducted with CDC miniature light traps (Model 512,

John W. Hock Company, Gainesville, FL) suspended 0.3m above the ground in the shaded edges of mosquito harborage zones (Fig. 6-2). The traps were baited with CO2 and skin lure (BG-Lure

™ cartridge, BioGents AG, Regensburg, Germany) and the light sources disabled for the duration of this study, based on methods by Holderman et al. (2018) to maximize inclusion of synanthropic mosquitoes. Two traps were placed within 6m of the vegetative midline of the study site, while an additional two traps were placed 60m away from the vegetative midline, within shaded harborages (Fig 6-2). Adult collections were made once per week for a continuous

24h sampling interval using a collection jar attached to the trap. After 24h, the trap, collection jar, and collected mosquitoes were removed to the laboratory and arthropods placed into storage containers. Trap units were repositioned the following week. Non-target species were separated from the contents of the jar and mosquitoes were identified to species within each collection.

Container-inhabiting mosquito surveillance was supplemented with 500ml black oviposition cups baited with oak infusion water prepared according to Trexler et al. (1998). Two baited oviposition cups were placed in the field continuously and were hidden in ground cover within 6m of the vegetative midline. An additional two cups were similarly hidden 60m away from the vegetative midline (Fig 6-2). To collect mosquito eggs, a seed germination paper

(#6512981311, Anchor Paper Co., Saint Paul, MN) was nested in each cup and partially submerged in the infusion water to provide and oviposition substrate. Papers were collected weekly and returned to the laboratory for egg enumeration under a dissection microscope. Adult and egg treatment surveillance was conducted for 4 weeks prior to the beginning of treatments to establish the pre-treatment mosquito abundance. Post-treatment surveillance continued for four weeks.

126

6.2.5 Treatment

Treatments were randomly assigned prior to the 4-week pre-treatment surveillance. After pre-treatment surveillance concluded, the vegetative midline of the treatment study sites were treated with the randomly assigned combination of products with a corresponding low or high dilution rates per 3,786 ml of water, or water-only control. The positive control was Onslaught

(6.4% esfenvalerate, 1.6% prallethrin, 8% PBO; Onslaught Fast Cap, McLaughlin Gormley King

Company, Minneapolis, MN) at 15ml and 30ml. The experimental treatment was metofluthrin at

30ml and 60ml. A final duplex treatment was made whereby Onslaught and metofluthrin were mixed in one tank at their equivalent low or high rates in 3,786 ml of water. Control sites were treated with only water. The aforementioned treatment combinations encompassed three comparisons for both low and high rates with an accompanying control for each set of treated sites, totaling six treated sites and two control sites per replicate, with three replicates conducted.

All barrier applications were made at a rate of 620ml per 50m2 using a battery powered backpack sprayer (REC 15 ABZ, Birchmeier Sprühtechnik AG, Stetten, Switzerland). Backpacks were set to a 600kPa flow pressure, fitted with the manufacturer supplied cone nozzle, and calibrated to a flow rate of 1,350 ml/min and droplet size volume median diameter (VmeD) and diameter of

50% droplet volume (Dv0.5) of 101–168 microns (101 µ ≤ Dv0.5 ≤ 168 µ). The vegetative midline of each treatment site received a 60-m transect of the insecticide application at a 3-m spray height and 6-m penetration depth, totaling 1,080m2 of treated area. Applications were performed by a licensed pesticide applicator between 8:00am and 10:00am while moving at 134 m/min.

6.2.6 Leaf Excision Bioassays

Persistence of pesticides applied to field sites was measured using leaf bioassays. Once per week, six leaves were excised at random from within 10m transects along the central barrier line from each treatment and control site. Sampled leaves had an upper surface area between 25–

127

30cm2. Upon return to the laboratory, each leaf was contained in a standard polystyrene

100×15mm Petri dish (#S33580A, ThermoFisher Scientific, Waltham, MA). For each bioassay, a non-blood-fed cohort of ten adult, female, 5-7d old Ae. albopictus were aspirated into each treatment and control Petri dish through a 20-mm opening (Fig 6-3). During the assay, Petri dishes were stored in rearing conditions with the opening sealed using cotton saturated with 10% sucrose solution. Knockdown, as indicated by ataxia, and mortality, as indicated by complete non-responsiveness to stimulus, was each scored after 20min and 24h of continuous exposure.

Leaf bioassays were repeated once a week for 4 weeks.

6.2.7 Data Analysis

For aging bioassays and leaf bioassays, the knockdown and mortality of Ae. albopictus was anlyzed using ANOVA and Tukey HSD test to compare treatments across each week.

Mosquito adult and egg surveillance was analyzed with a multi-way ANOVA and a multiple pairwise comparison with Tukey HSD using treatment type, the distance of the trap, and application rate as main effects for each of the weekly count of all adult mosquitoes, Ae. albopictus adults, and Ae. albopictus eggs. Full factorial interaction terms between treatment type, distance of the trap and application rate were also analyzed. The aforementioned analyses were carried out with JMP 13.1.0 (SAS Institute, Inc., Cary, NC). Percent reduction in the relative abundance of wild mosquitoes in the field, as measured by adult and egg surveillance,

% was calculated using Mulla’s formula: R = 100 ˗ [(C1/T1) × (T2/C2)] × 100; where C1 = pre- treatment measure of mosquito abundance in the associated control site, C2 = post-treatment mosquito abundance in the control site, T1 = pre-treatment mosquito abundance in the treated site, and T2 = post-treatment mosquito abundance in the treated site (Mulla et al. 1971).

128

6.3 Results

6.3.1 Residual Longevity

For aging bioassays, knockdown and mortality of assayed mosquitoes after exposure to aged metofluthrin can be visualized in Fig. 6-4. Following treatment, bioassay results during the week of treatment and the following week did not significantly differ across environmental conditions, excluding the control. During week 2, knockdown significantly decreased in all environmental conditions having greater than 80% humidity (L>80, D>80, L/D>80), the condition receiving simulated rainfall, and the unregulated outdoor condition (F8,26 = 18.0, p <

0.0001). However, mortality decreased significantly only in the outdoor condition (F8,26 = 302.2, p < 0.0001). By week 3, the rainfall simulated condition, outdoor, and D>80 conditions were no longer significantly different from the control, and the D<30 and L/D<30 conditions had significantly greater knockdown than all others (F8,26 = 29.4, p < 0.0001). In contrast, mortality in the outdoor condition was not different from the control during the third week, while D<30,

L/D<30, and L<30 were significantly higher than the remaining conditions (F8,26 = 32.3, p <

0.0001). At week 4, only the L<30 condition yielded bioassay results where knockdown (F8,26 =

7.9, p < 0.0001) and mortality (F8,26 = 4.7, p < 0.003) were significantly greater than the controls, but the mean of both responses fell below 80% (Fig. 6-4).

The average knockdown and mortality of leaf bioassayed mosquitoes are summarized by the post-treatment week in Fig. 6-5 for both low and high application rates. Regardless of rate, the knockdown and mortality from samples collected during the first week post-treatment were

100%, as compared to 0% from the controls (F3,11 = 35.52, p < 0.0001). Starting in week 2 through 4, the low rate did not result in significant knockdown. Mortality in the second week declined to 10-50% from the first week (F3,11 = 5.76, p < 0.0213), but remained similar between treatments. By the third week, knockdown and mortality in the metofluthrin treatment did not

129

differ from the control, while Onslaught and the blended treatment remained significantly different (Fig. 6-5; F3,11 = 25, p < 0.0002). By the fourth week, the metofluthrin and Onlsaught treatments failed to produce mortality, while the blended treatment generated 30% mortality

(F3,11 = 184, p < 0.0001).

In contrast to the low rate, the high rate maintained significant knockdown with the metofluthrin and the Onslaught + metofluthrin blend (F3,11 =15.31, p < 0.0011), as well as 100% mortality for all treatments until week 3 (Fig. 6-5). At week 3 and 4, knockdown was no longer observed. Mortality remained at 50% for blended treatment, 40% for Onslaught, and 30% for metofluthrin during week 3 (F3,11 = 14.85, p < 0.0012). By the final week of sampling, low mortality was observed in the blended treatment, while no mortality was present in the other treatments (Fig. 6-5).

Average weekly rainfall for sites prescribed to low rate treatments was 2.8cm, while the high rate treatment sites received a weekly average of 4.1cm. The plants identified and sampled for the leaf bioassays were: Asimina sp., Carpinus caroliniana, Cornus florida, Hydrangea quercifolia, Ilex vomitoria, Liquidambar styraciflua, Phyllostachys aurea, Prunus caroliniana,

Quercus laurifolia, Quercus nigra, Quercus virginiana, and Vitis rotundifolia. Treatment and control sites shared this heterogeneous plant composition (Fig. 6-2, 6-3).

6.3.2 Treatment Surveillance

Although data are presented primarily for Ae. albopictus, the general species composition at field sites included Ae. albopictus, Ae. atlanticus Dyar and Knab, Ae. fulvus pallens

(Wiedemann), Ae. infirmatus Dyar and Knab, Ae. taeniorhynchus (Wiedemann), Anopheles atropos Dyar and Knab, An. crucians Wiedemann, An. perplexans (Ludlow), An. punctipennis

(Say), An. quadrimaculatus Say, Coquillettidia perturbans (Walker), Culex erraticus (Dyar and

Knab), Cx. nigripalpus Theobald, Cx. quinquefasciatus Say, Cx. territans Walker, Culiseta

130

melanura (Coquillett), Mansonia dyari Belkin, Heinemann, and Page, Ms. Titillans (Walker),

Psorophora columbiae (Dyar and Knab), Ps. ferox (Humboldt), and Uranotaenia lowii

Theobald. Percent reduction in combined post-treatment mosquito collections from sites treated at the low rate applications are summarized in the appendices (Fig. A-2). Appendices also contain the corresponding low application rate data specifically for Ae. albopictus adults (Fig. A-

3) and Ae. albopictus eggs (Fig. A-3). Differences in Ae. albopictus relative abundance at high rate treatment sites were more apparent, and are included in Fig. 6-6 for Ae. albopictus adult and egg collections. In the interest of brevity, high application rate treatment results from the general mosquito population at the same sites can be found in the appendix (Fig. A-2).

Combined post-treatment mosquito collections (F3, 15 = 10.5906, p < 0.0001) and collections of Ae. albopictus adults (F3, 15 = 8.8181, p < 0.0001) were significantly affected by treatment type, with Onslaught + metofluthrin reducing collected adult mosquitoes more than

Onslaught or metofluthrin as individual treatments. Meanwhile, collections of Ae. albopictus eggs were significantly affected by treatment type (F3, 15 = 19.5649, p < 0.0001) and the trap location (F1, 15 = 19.0549, p < 0.0001), where collections at 6m were significantly less than collections at 60m (Fig. 6-6). For Ae. albopictus egg collections, metofluthrin, Onslaught Fast

Cap, and Onslaught + metofluthrin all reduced egg collections significantly during post- treatment but were insignificantly different between treatment types. The interaction of treatment type by application rate was significant for each of the combined mosquito collection (F3,15 = 4.4, p < 0.0381), Ae. albopictus adults (F3, 15 = 8.87, p < 0.0035), and Ae. albopictus eggs (F3,15 =

2.97, p < 0.0349). The remaining interactions between treatment types, trap locations at 6m or

60m, and application rate did not have a statistically significant effect on the dependant variables. Percent reduction reported here focuses on the high rate for simplicity.

131

Counterfactual analysis with Mulla’s formula on the net effect of pre-treatment and post- treatment data with both control and treatment sites yielded percent reductions that were interpreted separately from the model analysis. Percent reductions from the high rate metofluthrin treatment reduced adult Ae. albopictus adult collection by 60-80% for both 6m and

60m traps during the first two post-treatment weeks (Fig. 6-6), followed by a rise in the mosquito collections in the latter two weeks. In contrast, Onslaught Fast Cap reduced collection at the 60m trap by 60% for the first week, while collections at the 6m trap were unchanged from pre- treatment levels. During weeks two and three, there was a 40-70% reduction in collected adults at 6m traps and 30-60% reduction at 60m traps relative to pre-treatment levels. This treatment was not effective during week four. For the Onslaught + metofluthrin treatment, reductions in adult collections were observed in all post-treatment weeks, with ≥90% reduction in the first two weeks and a drop to a 40-60% reduction by the third and fourth week (Fig. 6-6).

The numbers of eggs collected in oviposition cups placed in the metofluthrin treatment were 60-100% lower in all traps from the treated area as compared to the control treatment for the first three weeks. During the fourth week, the 6m ovicups maintained a 40% reduction in egg collection, while egg numbers increased by 40% in the 60m cups (Fig. 6-6). Egg collections in the Onslaught Fast Cap treatment were 80% lower than pre-treatment at the 6m ovicups for three weeks before dimishing to a 40% reduction by the fourth week. A reduction in egg collections from distal cups was not observed in Onslaught-treated sites. As with the adult collections, 40-

80% fewer eggs than pre-treatment were collected in cups placed in the duplex treatment sites, regardless of cup distance (Fig. 6-6). The trends in relative Ae. albopictus abundance for both adults and eggs following low rate treatment are similar to the high rate with weaker overall reduction in mosquito abundance (Fig. A-3).

132

6.4 Discussion

Regarding longevity, the United States Environmental Protection Agency promotes guidelines that product adulticidal efficacy should persist above 95% (EPA 2009). Using this guideline with the ageing bioassay data, we can infer that the experimental metofluthrin formulation should be expected to last 2-3 weeks in most lighting and humidity conditions at

~26.6±1°C. However, regular rainfall, particularly if it exceeds 2cm per week, can jeopardize longevity. The fully exposed, outdoor condition was not in a shaded area, and essentially represents the maximum decay to be expected in the field in reasonable conditions. It should be noted, though, that barrier treatments are recommended to be strategically applied in shaded areas around mosquito harborage (Cilek 2008; Cilek and Hallmon 2008).

The average longevity of 2-3 weeks observed in the aging bioassays was echoed in the low rate results from the leaf bioassays, but observed differences can be attributed to average rainfall during field study being greater than what was simulated in aging bioassays. The average rainfall in the field was twice that of simulated conditions, which may explain rapid decline of observed knockdown effects during leaf bioassays. Data from high rate applications implies that the longevity of metofluthrin and Onslaught Fast Cap blended together is superior to either single product applications. Similarly, the data from field surveillance at the treated sites corroborated a greater percent reduction, and a greater sustained reduction across time, of mosquitoes and eggs.

In constructing the study, there was a suspicion based on the behavioral analysis from

Bibbs et al. (2019b) that mosquito numbers may increase at 60m trap locations as a consequence of metofluthrin having strong repellent action and that repellents have been incriminated in pushing mosquitoes to unprotected areas or hosts (Moore et al. 2007). An intervention using spatial repellents to abate malaria failed to see meaningful area-wide decline in infection rates, leading the authors to suggest that repelled vectors were diverted to unprotected hosts (Maia et

133

al. 2016). Thus, there were concerns that the aforementioned behavioral study (Bibbs et al.

2019b) documented metofluthrin to cause up to 60% repellency, which strongly overshadowed the 10% repellency caused by Onslaught Fast Cap (Bibbs et al. 2019b). Yet, adult surveillance data from the current study did not provide a clear indication of push effects of host seeking mosquitoes to the distal traps. Generally, the reductions in collection were similar between 6m and 60m locations. However, there were notable exceptions. In the general mosquito population, low rate applications of Onslaught failed to reduce mosquito collection at 60m traps (Fig. A-2).

Whereas metofluthrin and the duplex treatments resulted in observed reduction in both 6m and

60m traps in the early weeks of surveillance.

When focusing on Ae. albopictus eggs, Onslaught Fast Cap treatments did not result in significantly fewer eggs collected at 60m traps throughout the surveillance. The metofluthrin and duplex treatments were contrary in that fewer eggs were collected at both 6m and 60m traps throughout the study. Perhaps more importantly, the low rate application of Onslaught dramatically increased eggs collected after the first week and for the remaining collections. This persistent increase in eggs may be indicative of egg dumping behavior or a hormetic increase in egg yield as a result of pesticide-induced stress (Bong et al. 2017). Possible egg dumping was observed to a lesser degree after low rate treatments of the Onslaught + metofluthrin duplex during the second and fourth week at 60m traps. The increase in eggs collected at 60m traps accentuates that push effects are not limited to movement of the host seeking mosquitoes.

Unintentional pushing of gravid females would result in a population that becomes refractory to the control measure employed, which has been observed when source reduction results in gravid females moving to nearby areas (Unlu et al. 2013). However, prior study shows evidence that mosquitoes surviving exposure to metofluthrin do not necessarily oviposit viable eggs (Bibbs et

134

al. 2019a). We did not confirm in this study whether the eggs collected, despite increase or decrease in abundance, would hatch and develop without fitness detriments.

Overall, formulated metofluthrin as a standalone treatment should persist for 2-3 weeks in the field, even in the event of high rainfall. Residual treatments should meet at least a 4-week efficacy benchmark (Cilek 2008, Cilek and Hallmon 2008). Although abundance of mosquitoes collected did not always appear to be reduced to low levels, the leaf bioassay results indicate that the integrity of the insecticidal action was favorable for a duration that meets the expectation of an outdoor residual product. Furthermore, metofluthrin may have a stronger role being included as a supporting ingredient, similar to how prallethrin is incorporated into products such as Duet with the intention of flush mosquitoes out of harborage to improve overall control (Farajollahi et al. 2012, Gibson et al. 2016, Dye-Braumuller et al. 2017). Behavioral analysis has indicated that prallethrin lacks the volatility and desired impact on mosquitoes when trying to enhance adulticides for mosquito control (Dye-Braumuller et al. 2017).

Metofluthrin has the potential to fill the performance gap, but the blend of metofluthrin on top of prallethrin in the Onslaught Fast Cap may not clearly allow identification of metofluthrin’s merits as a secondary active ingredient. In future work, other products should be blended wth metofluthrin for a side-by-side comparison in the field. Data from a more extensive comparison may improve the scope of understanding for how metofluthrin may synergize outdoor residual treatments, or perhaps ULV treatments. Given the present data, it is clear that metofluthrin can succeed as a residual treatment when targeting peridomestic environments and there may be greater potential for synergizing formulations containing multiple active ingredients.

135

6.5 Disclosures

Funding for this research was provided in part by the Florida Department of Agriculture and Consumer Services: Florida Coordinating Council on Mosquito Control research subcommittee project 025365. McLoughlin Gormley King (MGK) Company and Sumitomo

Chemical Company provided formulated metofluthrin, Onslaught Fast Cap, and mixing instructions for this study. Research reported in this publication was supported by the University of Florida Clinical and Translational Science Institute, which is supported in part by the NIH

National Center for Advancing Translational Sciences under award number UL1TR001427. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

136

6.6 Tables

Table 6-1. Daily environmental conditions for aging metofluthrin treated substrates Temperature Relative Hours of Hours of Label Rainfall ml (min-max) (°C) Humidity (%) Light Dark

D≤30 26±1 25±5 0 24 0

D≥80 26±1 85±5 0 24 0

L≤30 26±1 25±5 24 0 0

L≥80 26±1 85±5 24 0 0

L/D≤30 26±1 25±5 12 12 0

L/D≥80 26±1 85±5 12 12 0

R2cm 26±1 85±5 12 12 1,200 (1,200-1,200) Outdoor 28±2 70±25 12±1 12±1 2,075 (360-5,400) Control 26±1 45±5 12 12 0

Condition R2cm involves rinsing substrates with 1200ml of water once per day to simulate 2cm of daily rainfall (USGS 2017) on 625cm2 substrates. Condition ‘Outdoor’ involves exposure to outdoor conditions, with rainfall and temperature recorded Control, no treatment or special conditions; ‘L’ represents 24hr lighting, ‘D’ represents 0hr lighting, ‘L/D’ represents 12hr lighting; <30 represents under 30% humidity, >80 represents over 80% humidity; ‘R2cm’ represents 2cm of simulated weekly rainfall; ‘Outdoor’ represents unregulated exposure to ambient outdoor temperature, humidity, and rainfall.

137

6.7 Figures

Figure 6-1. Environmental chambers for housing treated wood substrates inside incubators. Aluminum pans elevate the substrate to prevent direct contact with other materials. A) Humidification chamber filled with perlite and water. B) Dessication chamber filled with silica. C) Bioassay design displaying mosquito-contianing exposure cages placed on treated wood substrates to determine metofluthrin efficacy persistence across a weekly aging experiment.

138

Figure 6-2. Field sites used for application of barrier treatments. Sites consisted of vegetation suitable for treatment, mosquito harborage for various mosquito species, and shaded edges to place oviposition cups baited with oak infusion water and CDC miniature light traps (Trap) suspended 0.3m or lower from the ground, baited with CO2 and BG lure, without a light. A) Trap set within 6m of treated vegetation and adjacent to larval habitat. B) Trap set within 6m of treated vegetation between two residential parcels. C) Traps set 60m from treated vegetation and accompanied by an oviposition cup (within dashed white circle). D) Trap (left, within dashed white circle) and oviposition cup (right) set 60m from treated vegetation along the opposite facing parcel.

139

Figure 6-3. Leaf bioassays where six leaves were removed in 10m transects from the treated vegetation of every field site. Tests were conducted by aspirating 10 adult, 5-7d old female Aedes albopictus (Skuse) into holding dishes and supplying with 10% sucrose cotton. Knockdown was measured after 20min and mortality after 24hr.

140

Figure 6-4. Aging bioassay data visualizing percent knockdown (KD) after 20min or mortality after 24hr in adult female Aedes albopictus (Skuse), sorted by the week of bioassay and categorized according to environmental condition used to age meotfluthrin residual treatments. Environmental conditions include: Control, no treatment or special conditions; ‘L’ represents 24hr lighting, ‘D’ represents 0hr lighting, ‘L/D’ represents 12hr lighting; <30 represents under 30% humidity, >80 represents over 80% humidity; ‘R2cm’ represents 2cm of simulated weekly rainfall; ‘Outdoor’ represents unregulated exposure to ambient outdoor temperature, humidity, and rainfall. Bars represented with standard error of the mean as I- bars.

141

Figure 6-5. Leaf bioassay data visualizing the change in percent knockdown (KD) after 20min or mortality after 24hr exposure in adult female Aedes albopictus (Skuse), displayed by the post-treatment week of leaf bioassay and categorized according to treatment. Leaves were excised from vegetation at study sites having randomly assigned treatments with corresponding low (A) or high (B) dilution rates per 3,786 ml of water: Onslaught (6.4% esfenvalerate, 1.6% prallethrin, 8% PBO) at 15ml/30ml, metofluthrin (EXP141610001, 32% metofluthrin) at 30ml/60ml, or a duplex treatment containing both. Bars shown with standard error of the mean as I-bars. Average rainfall across the study was 2.8 cm and 4.1cm per week for low and high rate periods, respectively.

142

Figure 6-6. Percent reduction in the relative abundance of Aedes albopictus (Skuse) in the field, as measured by adult (A) and egg (B) % surveillance, calculated using Mulla’s formula: R = 100 ˗ [(C1/T1) × (T2/C2)] × 100; where C1 = pre-treatment measure of mosquito abundance in the associated control site, C2 = post-treatment mosquito abundance in the control site, T1 = pre- treatment mosquito abundance in the treated site, and T2 = post-treatment mosquito abundance in the treated site (Mulla et al. 1971). The corresponding percent reduction is shown with 95% confidence for traps both 6m and 60m away from vegetation treated with a high label rate for either metofluthrin, Onslaught Fast Cap, or a duplex treatment of Onslaught + metofluthrin. In some instances, population increased in a given week of surveillance, which is displayed in the red region below the 0 on the x-axis.

143

CHAPTER 7 EFFICACY OF METOFLUTHRIN IN RESIDUAL INSECTICIDE BLENDS FOR MOSQUITO CONTROL

7.1 Introduction

Volatile pyrethroids have been reviewed favorably as spatial repellents when implemented in various styles of emanating device (Achee et al. 2012a, Bibbs and Kaufman

2017). Despite the growing body of research on spatial repellency, various sources of evidence have led to the suggestion that the potential of volatile pyrethroids used as repellents may be squandered given the high comparative toxicity of several active ingredients, particularly nor- trans-chrysanthemate pyrethroids such as transfluthrin and metofluthrin (Bibbs et al. 2018a;

2018b; 2019a; 2019b) A wide array of sub-lethal fitness impacts were observed in mosquitoes exposed to transfluthrin (Bibbs et al. 2018a) and metofluthrin (Bibbs et al. 2019a), regardless of resistance phenotype among the mosquitoes (Bibbs et al. 2019a). These experiments support the inclusion of volatile pyrethroids in alternative delivery systems, given that consumer products are limited in practice (Bibbs et al. 2018a; 2018b; 2019a). Exploration of metofluthrin efficacy for operational mosquito management revealed that metofluthrin could be stabilized as a residual treatment and still elicit a spectrum of toxic effects as was observed against mosquitoes while they attempted to move upwind towards host cues (Bibbs et al. 2019b). The positive performance of the residually active metofluthrin in comparison to Onslaught Fast Cap, a commercially available alternative, ushered in a penultimate step of deploying metofluthrin as a fully formulated outdoor residual treatment for application to mosquito harborage sites (Ch. 6).

Incidentally, it was postulated that metofluthrin may have contributed towards mosquito reduction not only as a singular toxicant, but also as a supporting ingredient in products containing multiple active ingredients (Ch. 6).

144

The potential for synergy among volatile pyrethroids has been recognized in area-wide mosquito management (Farajollahi et al. 2012, Dye-Braumuller et al. 2017), namely with prallethrin as a supporting active ingredient in products such as Duet (sumithrin/prallethrin mixture). To date, prallethrin has been characterized as an irritant intended to encourage flight, and subsequently increased insecticide exposure risk, as mosquitoes abandon harborage sites

(Farajollahi et al. 2012). Pallethrin itself offers some contribution to the toxicity of a mixture

(Bibbs et al. 2018b), despite that it pales in comparison to the volatility of metofluthrin (Bibbs et al. 2018b). Specific investigation of prallethrin as a stimulant for enhancing adulticide application has demerited the compound as a secondary toxicant, resulting in a call to identify supporting molecules that have greater net effect than prallethrin (Dye-Braumuller et al. 2017).

Behavioral comparisons using a wind tunnel strongly support that metofluthrin overshadows prallethrin as a volatile secondary ingredient (Bibbs et al. 2019b). Furthermore, preliminary field data increases the evidence that products blended with metofluthrin may enhance management outcomes (Ch. 6). Continued work on synergy of toxicant blends containing metofluthrin could lead to better availability of tools similar to, or more effective than, prallethrin blended products.

Persistent efforts to understand the beneficial qualities of metofluthrin have targeted peridomestic, container-inhabiting mosquitoes as the model for understanding methods of delivery (Kawada et al. 2006, Ritchie and Devine 2013), ranges of toxic effects (Buhagiar et al.

2017a, Bibbs et al. 2018b; 2019) and general strategies of reducing vector contact with humans

(Kawada et al. 2005; 2006; 2008). This ubiquity is in part because of the difficulty of reducing the abundance of Aedes albopictus (Skuse) and Ae. aegypti (L.) across large areas (Faraji and

Unlu 2016). In kind, recent work keeps to this trend of focusing on peridomestic mosquitoes in an effort to clarify how best to use versatile tools such as metofluthrin (Bibbs et al. 2019b, Ch. 6)

145

for preventing urban pathogen transmission by Ae. albopictus and Ae. aegypti. However, despite the emphasis, metofluthrin is not limited to controlling Aedes (Stegomyia), as other vectors, such as Culex (Kawada et al. 2005) and Anopheles (Kawada et al. 2008) are impacted equally well by metofluthrin.

Therefore, the previous work with metofluthrin as a barrier product is an initial step towards understanding how best to implement volatile pyrethroids in delivery formats amenable for mosquito abatement programs. Synergistic blends of Onslaught Fast Cap with metofluthrin appeared to improve both longevity of efficacy and mosquito reduction (Ch. 6). Onslaught Fast

Cap includes prallethrin as a part of its product formulation confounding progress in determining if metofluthrin could be a successor to prallethrin as a secondary active ingredient. A more thorough comparison of residual products blended with metofluthrin needs to be conducted to examine synergy by metofluthrin in the field. It also was unclear if eggs collected from treated locations were viable (Ch. 6), given that prior work with Ae. aegypti and metofluthrin indicated a wide range of fitness impacts among mosquitoes surviving metofluthrin exposure (Bibbs et al.

2019a). Metofluthrin synergy may not be limited to the increased longevity of a blend; the observed improvements in population reduction may be owed in part to sub-lethal effects. In an expansion of Chapter 6, field assessment of metofluthrin was replicated using a larger selection of products with the specific intent to compare blends to determine if metofluthrin is a strong supporting ingredient. As a secondary objective, samples were screened for possible mosquito fitness changes among the various treatments to determine if sub-lethal effects documented in laboratory investigations (Bibbs et al. 2018a; 2019) could be detected in the field.

146

7.2 Materials and Methods

7.2.1 Insect Rearing

Eggs of Gainesville 1992 strain Ae. albopictus were obtained from the United States

Department of Agriculture, Center for Medical, Agricultural, and Veterinary Entomology in

Gainesville, FL and reared at the Anastasia Mosquito Control District in St. Augustine, FL for use during leaf excision bioassays. Subjects were reared at 26.6°C, 85±5% relative humidity

(RH), with a photoperiod of 14:10 (L:D). To replenish the colony, oviposition papers holding

2,000 eggs were submerged in larval pans filled with 2,500ml of reverse osmosis (RO) water.

Emerging larvae were fed 1-3g of liver and yeast mixture at a 3:2 ratio ad libitum in a 50-ml suspension. Adults were housed in flight cages and provided with separate supplies of 10% sucrose solution and RO water. All leaf bioassays were conducted using 5–7d old, non-blood- fed, female Ae. albopictus.

7.2.2 Site Evaluation

Criteria for field sites and associated surveillance were replicated from Chapter 6. In brief, residential parcels in St. Augustine, FL that were separated by a 1km buffer and containing approximating 0.8ha areas were selected based on the availability of vegetative harborage and oviposition resources for peridomestic mosquitoes. Four miniature CDC ‘no-light’ traps were deployed at each site for weekly 24h trap nights according to methods from Chapter 6 and previously adapted from Holderman et al. (2018). Four accompanying 500ml black oviposition cups lined with an oviposition substrate and filled with oak infusion water (Trexler et al. 1998) were nested at the field sites for weekly collection. The array of traps spanned a 6m set of traps proximal to mosquito harborage and a 60m set of distal traps with respect to the same harborage.

Mosquito populations, focusing on Ae. albopictus, were monitored for 4 weeks prior to treatment and for 4 weeks following treatment. A total of 18 sites were monitored: 8 sites were designated

147

for low rate treatment with 1 accompanying control, while a second set of 8 treatment sites were allocated for high rate with 1 accompanying control. A 32% metofluthrin formulation

(EXP141610001, McLaughlin Gormley King Company, Minneapolis, MN) was selected for comparison with Onslaught Fast Cap (6.4% esfenvalerate, 1.6% prallethrin, 8% PBO), Hyperion

(10% d-phenothrin and 10% PBO), and Sector (10% permethrin and 10% PBO). Three different duplex treatments were used whereby metofluthrin was added to each of the aforementioned three products. As a form of positive control, OneGuard (0.4% prallethrin, 1.3% pyriproxyfen,

4% lambdacyhalothrin, and 6% PBO) was tested as an example of a volatile residual intended to manage mosquitoes at all life stages. Treatmeants were replicated three times at these sites with a

4-week gap between treatments to allow mosquito numbers to recover before subsequent replications.

After the pre-treatment surveillance period, applications to the vegetative midline of the treatment study sites were made with randomly assigned product mixtures with corresponding low or high dilution rates per 3,786 ml of water: OneGuard mixed at 30ml/60ml, Onslaught at

15ml/30ml, Sector at 96ml/192ml, Hyperion at 15ml/30ml, and metofluthrin at 30ml/60ml. For duplex treatments, Onslaught, Sector, and Hyperion were each mixed with the corresponding low or high rate for metofluthrin when diluted in 3,786 ml of water. Control sites were treated with only water. All barrier applications were conducted at the same specifications as Chapter 6.

Leaf bioassays were replicated in procedure from Chapter 6, but for all eight treatment combinations and untreated control. In brief, once per surveillance week, six leaves were excised at 10m transects along the central barrier line from each treatment and control site. Leaves were placed individually into marked Petri dishes that received 10 non-blood-fed, 5-7d old, adult,

148

female Ae. albopictus and 10% sucrose supply. Knockdown after 20min and mortality after 24h continuous exposure were recorded and analyzed.

7.2.3 Field Sample Rearing

Operating under the assumption that high rate field treatments using metofluthrin were more likely to produce noticeable trends (Ch. 6), all eggs collected from the high rate sites (eight allocated to treatment and one for control) were reared in the laboratory. Oviposition papers that were removed from the field were labeled as originating from a 6m or 60m sampling location.

After eggs were enumerated for surveillance purposes, they were stored for 24h in the insect rearing conditions described earlier. A visual inspection was made after the 24h holding period before eggs were submerged in 350ml of RO water held within a previously unused 500ml black oviposition cup, in same likeness to those used for field surveillance. Emergent larvae were reared in the insectary rearing conditions using the respective cups until pupation, with food provided ad libitum. After 7d submerged in water, oviposition papers were inspected again to confirm any observations made after the 24h-holding period. Eggs with visual deformities, such as incomplete melanization or collapsed chorion, were tallied for each cup and removed from the total of viable eggs recorded for the treatment.

7.2.4 Data Analysis

For each of knockdown and mortality from leaf bioassays and the percent of viable eggs from the field, a one-way ANOVA and Tukey’s HSD was used to compare insecticide application treatments. For each of the combined mosquito totals, Ae. albopictus adults, and Ae. albopictus eggs, collected in CDC traps and oviposition traps, respectively, a multi-way

ANOVA and a multiple pairwise comparison with Tukey HSD test were conducted using treatment type, distance of the trap, and application rate as main effects. Interaction terms between the main effects were included in the analysis. Statistical analysis was performed in

149

JMP 13.1.0 (SAS institute, Inc., Cary, NC). Percent reduction following treatments was

% calculated using Mulla’s formula: R = 100 ˗ [(C1/T1) × (T2/C2)] × 100; where C1 = pre- treatment measure of mosquito abundance in the associated control site, C2 = post-treatment mosquito abundance in the control site, T1 = pre-treatment mosquito abundance in the treated site, and T2 = post-treatment mosquito abundance in the treated site (Mulla et al. 1971).

7.3 Results

7.3.1 Field Surveillance

Trends in the general mosquito population closely mirrored findings with Ae. albopictus adults. Consequently, our results will focus on Ae. albopictus as a model, but the results with the general mosquito population are summarized in the appendix (Fig. A-4). The species composition included Ae. albopictus, Ae. atlanticus Dyar and Knab, Ae. fulvus pallens

(Wiedemann), Ae. infirmatus Dyar and Knab, Ae. taeniorhynchus (Wiedemann), Ae. triseriatus

(Say), Anopheles atropos Dyar and Knab, An. crucians Wiedemann, An. perplexans (Ludlow),

An. punctipennis (Say), An. quadrimaculatus Say, Coquillettidia perturbans (Walker), Culex erraticus (Dyar and Knab), Cx. nigripalpus Theobald, Cx. quinquefasciatus Say, Cx. territans

Walker, Culiseta melanura (Coquillett), Mansonia dyari Belkin, Heinemann, and Page, Ms. titillans (Walker), Psorophora columbiae (Dyar and Knab), Ps. ferox (Humboldt), Ps. howardii

Coquillett, Uranotaenia lowii Theobald, Ur. Sapphirina (Osten Sacken), and Wyeomyia mitchelli

(Theobald).

Treatment type was a significant factor influencing the combined post-treatment adult mosquito collections (F8, 35 = 13.35, p < 0.0001) and the Ae. albopictus adult collections (F8, 35 =

11.62, p < 0.0001). Collections of Ae. albopictus eggs were influenced significantly by both treatment type (F8, 35 = 6.27, p < 0.0001) and the distance of the oviposition traps (F8, 35 = 11.45, p < 0.0009), with 60m traps generally have more mosquitoes than 6m traps (t = 3.39, p <

150

0.0009). The interaction between treatment type and application rate was significant for combined mosquito collections (F1, 8 = 42.02, p < 0.0001), Ae. albopictus adults (F1, 8 = 36.61, p

< 0.0001), and Ae. albopictus eggs (F1, 8 = 8.87, p < 0.0032). The remaining interactions between treatment type, distance of the traps, and rate were not significant. Pairwise comparisons between treatment types supported that Hyperion + metofluthrin, Sector + metofluthrin, and Onslaught + metofluthrin treatments reduced Ae. albopictus adult and egg collections to a greater extent than the unblended products, though all treatments resulted in fewer mosquitoes collected than the control sites. The low rate application data are summarized in the appendix (Fig. A-5) for Ae. albopictus adult and egg surveillance, however it will not be discussed in the main passage of this study in light of the greater significance of treatments made at the high rate (adults: t = 6.05, p < 0.0001), eggs: t = 2.98, p < 0.0032).

Generally, post-treatment collections of adult Ae. albopictus increased after treatment with either Sector or Hyperion, with 60m traps collecting up to twice as many mosquitoes as prior to treatment (Fig. 7-1). Onslaught Fast Cap reduced adult mosquitoes by 60-80% at 6m traps and 40-60% reduction at 60m traps (Fig. 7-1). Metofluthrin treatment resulted in 50-80% reduction at both trap distances for the first two weeks, but otherwise did not reduce the number of collected mosquitoes during weeks 3 and 4. Sector + metofluthrin provided 60-80% reduction in adult Ae. albopictus for three weeks before failing at the fourth week. Hyperion + metofluthrin provided closer to 90% reduction during the first three weeks, primarily at the 6m trap sites, before mosquito collections were no longer reduced during the fourth week. Onslaught + metofluthrin and OneGuard were the only treatments to reduce Ae. albopictus adults consistently by 60-80% for the entire post-treatment surveillance period (Fig. 7-1).

151

Egg data supports the general trends observed with adults, but with a few key differences

(Fig. 7-1). Sector had 70% reduction in eggs at the 60m trap for the first week, but otherwise did not result in fewer collected eggs at either trap distance for the rest of the study. Sector + metofluthrin and Hyperion resulted in 40-60% fewer eggs collected at 6m traps for 2 weeks, but this was accompanied by more than double the number of Ae. albopictus eggs collected at 60m traps. Onslaught Fast Cap resulted in 80-90% fewer eggs at 6m traps, but did not reduce egg collections at 60m traps (Fig. 7-1). Metofluthrin reduced egg collections by at least 50% at both sets of traps for three weeks, with diminished effect at the 60m trap in the last week. In contrast with the singular treatment, Onslaught + metofluthrin resulted in 40-80% fewer collected eggs throughout the study (Fig. 7-1). OneGuard and Hyperion + metofluthrin treatments resulted in what appeared to be a temporary egg dumping response, with more than double the number of eggs collected, but only during the first week. For OneGuard, all 6m collections had an 80-100% reduction in eggs each week, while 60m collections were reduced by 60% for all weeks after the first. Hyperion + metofluthrin had the strongest effect, with 100% reduction in all egg collections, regardless of distance, for the second and third weeks and 60-80% reduction at the

6m traps for the first and fourth week (Fig. 7-1).

7.3.2 Leaf Bioassays

The efficacy of residues on leaves tested in bioassays against Ae. albopictus is summarized in Fig. 7-2. Insecticide treatments could be grouped by similar efficacy relative to results from untreated controls. When comparing low rate applications, the first week of bioassays resulted in metofluthrin and all treatments blended with metofluthrin having 60-80% knockdown (F8,26 = 8.73, p < 0.0001). OneGuard, Hyperion, and Onslaught Fast cap yielded significantly less knockdown, while Sector was negligibly different from the control. During the

152

second week, the Hyperion + metofluthrin was the only treatment which retained 60-80% knockdown, while all other treatments effectively lost their knockdown effect (F8,26 = 6.56, p <

0.0005). Knockdown dissipated in all treatments for the remaining weeks. In contrast, the high rate resulted in 100% knockdown during the first week in all treatments, while during the second week OneGuard, Sector + Metofluthrin, Onslaught + metofluthrin, and metofluthrin by itself still produced high knockdown in spite of loss of knockdown in the other treatments (F8,26 = 71.43, p

< 0.0001). The remaining weeks, as with the low rate, essentially did not produce knockdown

(Fig 7-2).

High rate treatments differed greatly, with OneGuard being the only treatment to produce

100% mortality for three weeks (Fig. 7-2). All other treatments produced 100% mortality during the first week alone. Onslaught Fast Cap, metofluthrin, and all treatments blended with metofluthrin produced ~80% or greater mortality for the first two weeks (F8,26 = 86.35, p <

0.0001). During the third and fourth week, Sector + metofluthrin and Onslaught + metofluthrin continued to produce 50-80% mortality (F8,26 = 19.31, p < 0.0001), while all other treatments fell below 20%. Sector + metofluthrin, Hyperion + metofluthrin and Onslaught + metofluthrin remained above 90% mortality for the second week, with Onslaught and metofluthrin producing

50-80% mortality (F8,26 = 71.43, p < 0.0001). Sector + metofluthrin, followed by Hyperion + metofluthrin, Onslaught + metofluthrin, Onslaught, and metofluthrin continued to produce mortality above 40%, while all other treatments failed (F8,26 = 84.8, p < 0.0001). All treatments failed during the final week, possibly because of high rainfall.

Average weekly rainfall for sites prescribed to low rate treatments was 3.1cm with a range of 1.6-3.9cm, while the high rate treatment sites received a weekly average of 4.9cm with a range of 1.3-12.7cm. The max rainfall, 12.7cm at high rate sites, is incriminated for the general

153

failure of treatments during the last week of study, whether evaluating by leaf bioassay or reduction in mosquito collections. The plants identified and sampled for the leaf bioassays were:

Asimina sp., Bidens frondosa, Carpinus caroliniana, Carya glabra, Citrus sp., Cornus florida,

Hydrangea quercifolia, Ilex vomitoria, Liquidambar styraciflua, Magnolia grandiflora,

Phyllostachys aurea, Prunus caroliniana, Quercus laurifolia, Quercus nigra, Quercus virginiana, Rhododendron austrinum, Rhododendron canescens, Sambucus canadensis, and Vitis rotundifolia. Treatment and control sites generally shared this heterogeneous plant composition.

7.3.3 Field Sample Rearing

Only oviposition papers from field sites allocated for high rate treatment were used to conduct rearing studies. Visual inspection of the papers upon removal from the field and again after 24h in the laboratory confirmed that multiple egg phenotypes were present. Incompletely melanized, rubbery textured eggs were collected from all field sites treated with metofluthrin or a product blended with metofluthrin (Fig. 7-3A, 7-3B). Occasionally, eggs exhibited a collapsed chorion within the initial and 24h inspections (Fig. 7-3C). More often, eggs appeared collapsed and deformed following attempted rearing to confirm viability (Fig. 7-3D). Collapsed eggs were observed on 75% of ovipapers retrieved from field sites treated with metofluthrin or a product blended with metofluthrin. Egg collections from the control site and sites treated with OneGuard,

Sector, Hyperion, or Onslaught Fast Cap almost never resulted in collapsed eggs with less than

2% of egg papers containing 1-5 collapsed eggs.

The net effect of teneral and collapsed eggs is summarized as a mean percent reduction on the viability of collected Ae. albopictus eggs in Fig. 7-4. Once averaged across the four-week post-treatment surveillance window, the control, OneGuard, Sector, Hyperion, and Onslaught

Fast Cap sites effectively had no change in viability of the collected eggs (confirmed through

154

rearing). Sector + metofluthrin, Onslaught + metofluthrin, and metofluthrin treatments resulted in a reduction in viability of 20%, 30-40%, and 30% respectively. Hyperion + metofluthrin caused the highest reduction, 40%-55%, and significantly differed from the rest across proximal

(F8,49 = 9.71, p < 0.0001) and distal collection sites (Fig. 7-4; F8,56 = 36.2, p < 0.0001). There were no remarkable distinctions between samples from ovicups either proximal or distal to the treated harborage. Rearing ceased after larvae pupated; although OneGuard contains pyriproxyfen, quantifying emergence inhibition was not within the scope of this study.

7.4 Discussion

Field evaluation supports that metofluthrin enhances existing adulticide formulations.

Most metofluthrin blends used in this study became longer lasting than their singular products and resulted in greater suppression of both adult mosquitoes and eggs. Sector, a permethrin product, performed poorly even when blended with metofluthrin, particularly when compared with the increased longevity and control observed when blending metofluthrin with Hyperion and Onslaught Fast Cap. Although metofluthrin suppressed Ae. albopictus numbers when used by itself, the blended products demonstrate greater efficacy than unblended counterparts, particularly in the case of Hyperion + metofluthrin and Onslaught + metofluthrin. Similar improvements in adulticidal efficacy have been noted in studies using ULV applications during which Duet (sumithrin/prallethrin) reduced Ae. albopictus collections two-fold more than a non- volatile counterpart, Anvil (sumithrin) (Unlu et al. 2018). Despite field synergy when using prallethrin as an excitatory ingredient against Ae. albopictus (Unlu et al. 2018), there are conflicting investigations showing that prallethrin failed to produce significant differences under a tightly controlled setting (Dye-Braumuller et al. 2017). The behavioral analysis in which prallethrin failed was attributed to prallethrin demonstrating poor volatility (Dye-Braumuller et

155

al. 2017). The data on blends from this study supports that metofluthrin may be an option for a similar application methodology.

Additionally, the aforementioned blends resulted in the highest proportion of non-viable egg phenotypes collected from the field. Although the post-treatment surveillance showed 80-

100% reduction in Ae. albopictus adults and eggs following treatment with Hyperion + metofluthrin or Onslaught + metofluthrin, it is likely the damage to the local population are more distinct given the additional 40-50% reduction in viability of the collected eggs. The effect is difficult to interpret, as supporting evidence is lacking for Ae. albopictus sub-lethal effects.

Previous findings by Bibbs et al. (2018a) did not suggest that Ae. albopictus was prone to egg collapse as a result of sub-lethal exposure to volatile pyrethroids, in this case transfluthrin, especially when compared to Ae. aegypti during the same study. Follow-up work on a field strain of Ae. aegypti did show reduced viability in eggs from collapse, incomplete melanization, and premature hatching (Bibbs et al. 2019a). However, the lack of evidence specific to Ae. albopictus, particularly regarding field variants, makes it difficult to accurately interpret the teneral and collapsed eggs observed from the ovipapers retrieved during regular surveillance.

The leaf bioassay data implies that the concentration of toxicant on the leaves degrades quickly, as evidenced by the quick abatement of knockdown. The decline of knockdown effects has been used as a proxy in study of leaf residuals to determine when the toxicant is waning in efficacy (Allan et al. 2009). The leaf bioassay data may be better interpreted as a pass/fail indicator of the leaves retaining insecticide. However, the loss of knockdown within a week may indirectly point to an increased likelihood of sub-lethal exposure for the target mosquitoes. Loss of knockdown effect has been directly linked with the probability that residues are no longer acutely toxic to mosquitoes (Cilek 2008, Cilek and Hallmon 2008, Allan et al. 2009). Sub-lethal

156

metofluthrin exposure was severely detrimental to laboratory, field, and resistant phenotypes of

Ae. aegypti (Bibbs et al. 2019a). Although fitness decline is difficult to interpret with Ae. albopictus, it can be considered a bonus when making decisions based on the other measures.

Unlike in Chapter 6, there were strong push trends as a result of some treatments, particularly with egg collections. OneGuard, Onslaught + metofluthrin, Sector, Hyperion, and

Onslaught all showed evidence that mosquitoes would deposit eggs farther away from the immediate treatment area. However, post-treatment surveillance still showed that both 6m and

60m trap sites returned similar changes in relative abundance of adult mosquitoes. It is possible that surveillance would need to cover a larger total area, with respect to a 1km dispersal range of container-inhabiting mosquitoes (Honório et al. 2003), in order to properly discriminate push or pull effects. Aedes (Stegomyia) may generally navigate within 20m (Honório et al. 2003) or as much as 80m during skip oviposition (Davis et al. 2016), but the added stress of insecticide may prompt greater dispersal.

The only competition within this study group to products blended with metofluthrin was

OneGuard, a product intended to control all life stages of mosquito with an IGR, type-I volatile pyrethroid, type-II contact adulticide pyrethroid, and a PBO synergist. Although not prioritized in these results, low rate treatments indicated that OneGuard was ineffective when not used at a high rate (Fig. A-5). Metofluthrin blended treatments did not perform as poorly at the low rate against either the general mosquito population (Fig. A-4) or with eggs and adults of Ae. albopictus (Fig. A-5). Overall, the inclusion of metofluthrin strengthens previously unblended products and otherwise does not detract from the persistence of the base product. In addition, the high volatility of metofluthrin, even when formulated as a residual (Bibbs et al. 2019b), and the sub-lethal effects that come with the exposure to volatiles (Bibbs et al. 2019a) lay a foundation

157

for metofluthrin being a strong secondary ingredient. Metofluthrin blends not only perform well as an adulticide (Ch. 6), but potentially replace prallethrin as a preferred toxic synergist while providing a range of additional benefits. Volatile pyrethroids can have a powerful contribution to mosquito abatement programs, and henceforth studies should focus on creative methods of implementing volatile pyrethroids to best fit the needs of integrated vector management.

7.5 Disclosures

Chemical Company provided formulated metofluthrin, Onslaught Fast Cap, Hyperion, Sector,

OneGuard, and mixing instructions for this study. Research reported in this publication was supported by the University of Florida Clinical and Translational Science Institute, which is supported in part by the NIH National Center for Advancing Translational Sciences under award number UL1TR001427. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

158

7.6 Figures

159

Figure 7-1. Percent reduction in the relative abundance of Aedes albopictus (Skuse) in the field, as measured by adult (A) and egg (B) % surveillance, calculated using Mulla’s formula: R = 100 ˗ [(C1/T1) × (T2/C2)] × 100; where C1 = pre-treatment measure of mosquito abundance in the associated control site, C2 = post-treatment mosquito abundance in the control site, T1 = pre- treatment mosquito abundance in the treated site, and T2 = post-treatment mosquito abundance in the treated site (Mulla et al. 1971). The corresponding percent reduction is shown with 95% confidence for traps both 6m and 60m away from vegetation treated with a high label rate of either OneGuard, Sector, Hyperion, Onslaught Fast Cap, metofluthrin, or a duplex treatment of Sector/Hyperion/Onslaught + metofluthrin. In some instances, population increased in a given week of surveillance, which is displayed in the red region below the 0 on the x-axis.

160

161

Figure 7-2. Leaf bioassay graphs visualizing percent knockdown after 20min or mortality after 24hr in adult female Aedes albopictus (Skuse), sorted by the week of leaf bioassay and categorized according to treatment conditions. Leaves were excised from vegetation at study sites having randomly assigned product mixtures with corresponding low (A) or high (B) dilution rates per 3,786 ml of water: OneGuard (0.4% prallethrin, 1.3% pyriproxyfen, 4% lambdacyhalothrin, and 6% PBO) mixed at 30ml/60ml, Onslaught (6.4% esfenvalerate, 1.6% prallethrin, 8% PBO) at 15ml/30ml, Sector (10% permethrin and 10% PBO) at 96ml/192ml, Hyperion (10% d-phenothrin and 10% PBO) at 15ml/30ml, and metofluthrin (EXP141610001, 32% metofluthrin) at 30ml/60ml. For duplex treatments, Sector, Hyperion, and Onslaught were each mixed with the corresponding low or high rate for metofluthrin when diluted in 3,786 ml of water. Bars shown with standard error of the mean as I-bars. Average rainfall across the study was 4.1cm per week.

162

163

Figure 7-3. Aedes albopictus (Skuse) eggs diagnosed with reduced viability after removal from field sites treated with residual insecticide. A) Normal phenotype egg paired with an incompletely melanized, teneral egg immediately after the substrate was removed from a field site treated with a blend of Onslaught Fast Cap (esfenvalerate, prallethrin, PBO) and metofluthrin. B) Several teneral eggs remain after a 24h holding period in the laboratory (encircled with dashed outlines); the substrate was collected from a field site treated with a blend of Hyperion (sumithrin, PBO) and metofluthrin. C) Example of collapsed eggs, showing dimpled sides and accentuated keels, after a 24h holding period in the laboratory once the substrate was removed from a field site treated with only metofluthrin. D) Substrate containing only collapsed eggs after attempting to rear larvae for 7d from the eggs contained on the substrate; ovipaper was removed from a field site treated with a blend of Hyperion and metofluthrin.

164

Figure 7-4. Reduction of Aedes albopictus (Skuse) viable eggs from field sites treated with residual insecticide containing metofluthrin. Reduction in viability was determined by subtracting teneral and collapsed eggs from the total, confirmed by rearing the contents of the oviposition substrate after collection, and then averaging the findings across four post-treatment weeks. Summary is presented as mean percentages, grouped by the sample proximity to treated vegetation, with standard error of the mean as I-bars.

165

CHAPTER 8 CONCLUSIONS ON THE FUTURE OF VOLATILE PYRETHROIDS FOR INTEGRATED VECTOR MANAGEMENT

Spatial repellents have a deep cultural relevance among humans, with examples of many volatile compounds being used for managing biting insects throughout history (Rawlinson 1996).

The modern equivalents of burning repellent plants have manifested as coils (Xue et al. 2012a;

2012b), mats (Bibbs et al. 2015), vaporizers (Bibbs and Xue 2015), lamps (Shen et al. 2017), and various other emanators that dispense volatile compounds useful for keeping mosquitoes away

(Cook et al. 2007). Among those, volatile pyrethroids, particularly of the nor-trans- chrysanthemate subclass, have displayed a unique and potent affinity for killing mosquitoes

(Bibbs et al. 2018b), reducing fecundity (Bibbs et al. 2018a) or outright damaging fitness (Bibbs et al. 2019a, Ch. 7), and potentially combating insecticide resistance (Bibbs et al. 2019a). The expanded research on the potency of volatile pyrethroids, such as metofluthrin, have led to the understanding that they function well as adulticides (Ch. 6). Perhaps more importantly, they function well as supporting molecules because of high volatility that provokes strong behavioral response in mosquitoes (Bibbs et al. 2019b) while also improving the longevity and mosquito suppression power of multi-toxicant blends (Ch. 7). Ultimately, there is now exists a defensible body of work that demonstrates volatile pyrethroids offer a considerable range of benefits that can be successfully used in operational mosquito control.

Since volatile pyrethroids improve the management of mosquitoes on multiple fronts

(Bibbs et al. 2019a; 2019b, Ch. 6-7), a prudent development would be to include good candidates into combination products in the same likeness as prallethrin. Although there is an understanding that prallethrin facilitates management by agitating mosquitoes into risky flight behavior

(Farajollahi et al. 2012), it is neither the most toxic (Bibbs et al. 2018b) nor most volatile (Dye- 166

Braumuller et al. 2017) supporting ingredient available for that purpose. The success of including volatile pyrethroids into residual insecticides (Ch. 6, Ch. 7) is only one example of an effective delivery format that is amenable to mosquito abatement programs. The obvious transition would be towards ultra-low volume (ULV) insecticide formulations.

Although residual products can contain volatile secondary ingredients, such as Onslaught

Fast Cap (esfenvalerate/prallethrin), there is a larger body of work supporting ULV adulticide treatments with Duet (sumithrin/prallethrin) (Farajollahi et al. 2012, Unlu et al. 2014, Faraji and

Unlu 2016) The value of prallethrin in a ULV insecticide blend has been contentious, with firm examples of poor effect (Dye-Braumuller et al. 2017) and droplet analyses showing that spray droplets can physically penetrate confounding mosquito harborage (Faraji et al. 2016), thus dispelling ideas that prallethrin is the sole reason that mosquitoes are being reached with complex product blends. The principle behind prallethrin, or any blend containing a volatile excitatory agent, is a good idea: reach mosquitoes where droplets do not penetrate easily, agitate the mosquitoes to flush them out of their harborage, and consequently expose them to more toxic droplets containing another toxicant. The best route for some nor-trans-chrysanthemate pyrethroids may be to usurp the role of prallethrin and provide more potent combinations using the same concept. Behavioral analysis subjecting mosquitoes to vapors of metofluthrin versus a prallethrin blend gave ample evidence that at least metofluthrin could be a good candidate for such a process (Bibbs et al. 2019b).

The idea of using vapors to create a dual-exposure method to reach mosquitoes was not necessarily predicated on prallethrin. Aerial adulticiding with Dibrom, an organophosphate, uses the same principle to create a fast acting, rapid half-life treatment (Tietze et al. 1996) that has become highly preferred in mosquito control aerial operations (Rey 2014, Britch et al. 2018).

167

Dibrom contains naled, an insecticidal liquid, which decomposes into dichlorvos, a low residue insecticidal vapor, (Chen 1984) often during application (Hall et al. 1997). Although untested, the success of ground operational use may warrant the possibility for aerial application given that volatile pyrethroids have similar vapor initiating tendencies. At present, Dibrom is one of the few reliable aerial adulticides for mosquito control (Britch et al. 2018), but Dibrom’s status as one of the last organophosphates still registered for use in the United States has led to concerns about its future availability. Transfluthrin may be a candidate for aerial application as the vapors exhibit a heavier-than-air dispersal behavior (Jiang et al. 2019), suggesting that it may favorably descend during atmospheric drift during an aerial adulticiding application. Volatile pyrethroids could potentially shift the bias that exists for organophosphate aerial adulticides (Britch et al. 2018), contribute a tool to increase aerial versatility, and ensure that aerial adulticiding would not be crippled by the loss of Dibrom, should it ever occur.

Mosquito management receives more attention than management efforts of other vectors of public health importance. Mosquitoes are a suitable model for public health entomology, in part due to their ubiquity and ease of use during study. However, mosquitoes are far from singular as an important global public health concern. Ticks also are growing as a health threat, even despite that mosquitoes occupy the spot light. Outbreak control and population reduction for ticks suffers similar pesticide class restrictions as mosquitoes, and tick management has limited guidance on area-wide management compared to other vectors (Stafford 2007). Tick management relies heavily on avoidance, prevention, and early detection (CDC 2018). Some evidence suggests that volatile pyrethroids can affect ticks, at least as a repellent (Bibbs et al.

2016). Possibly, they could be a meaningful adulticide at close range (Bibbs et al. 2016).

Integration of spatial repellents is a potential option for host-focused adulticide strategies, similar

168

to how ectoparasitic pour-ons and feed-throughs are used in livestock and domestic animals.

Many hurdles exist for adding ticks to product labels, so extensive foundational research would need to be conducted towards this end. For example, what are the comparative toxicities and range of effect for volatile compounds against ticks? Are there ways to use volatile emanators to create barriers to tick entry through an area? Do pyrethroid resistance phenotypes negate the viability of volatile pyrethroids on this front? These questions are the minimum that would be posed against spatial repellents during future development in tick management.

Several potential research topics could be viable for volatile pyrethroids, or arguably spatial repellents from other chemical classes. In light of insecticide resistance, global movement of organisms, and the reticulated intersection of nature with humans, creative implementation of existing products, toxicants, and vector abatement strategies is the key to preserving the integrity of what few tools remain available for public health vector management. Sub-lethal effects, behavioral modification, multi-modal active ingredients, synergy, and duplex strategies are just part of the findings of this dissertation. There may yet be other volatile compounds that come from novel or sparsely used chemical classes (Gross et al. 2015; 2017, Norris et al. 2015) or expand our knowledge of synergy (Gross et al. 2017, Norris et al. 2018). Volatile pyrethroids are merely a beginning to the exploration of numerous paths for future integrated vector management.

169

APPENDIX

Figure A-1. Spectrum recording for broad spectrum ultraviolet lights used to decontaminate the modular wind tunnel of spatial repellent deposits. Spectrum was collected using a 350mA power input, 30cm away from the surface of the bulb, in a 23.89°C environment. The addition of ozone gas was not necessary to purge metofluthrin from the modular wind tunnel in part because of the partial UVC range in the emission spectra of the UV bulb. 170

Figure A-2. Percent reduction in the relative abundance of the general host-seeking mosquito population in the field, as measured by % adult and egg surveillance, calculated using Mulla’s formula: R = 100 ˗ [(C1/T1) × (T2/C2)] × 100; where C1 = pre- treatment measure of mosquito abundance in the associated control site, C2 = post-treatment mosquito abundance in the control site, T1 = pre-treatment mosquito abundance in the treated site, and T2 = post-treatment mosquito abundance in the treated site (Mulla et al. 1971). The corresponding percent reduction is shown with 95% confidence for traps both 6m and 60m away from vegetation treated with either a low (A) or high (B) rate of metofluthrin, Onslaught Fast Cap, or a duplex treatment of Onslaught + metofluthrin. In some instances, population increased in a given week of surveillance, which is displayed in the red region below the 0 on the x-axis. 171

Figure A-3. Percent reduction in the relative abundance of Aedes albopictus (Skuse) in the field, as measured by adult (A) and egg (B) % surveillance, calculated using Mulla’s formula: R = 100 ˗ [(C1/T1) × (T2/C2)] × 100; where C1 = pre-treatment measure of mosquito abundance in the associated control site, C2 = post-treatment mosquito abundance in the control site, T1 = pre- treatment mosquito abundance in the treated site, and T2 = post-treatment mosquito abundance in the treated site (Mulla et al. 1971). The corresponding percent reduction is shown with 95% confidence for traps both 6m and 60m away from vegetation treated with a low label rate of either metofluthrin, Onslaught Fast Cap, or a duplex treatment of Onslaught + metofluthrin. In some instances, population increased in a given week of surveillance, which is displayed in the red region below the 0 on the x-axis. 172

173

Figure A-4. Percent reduction in the general mosquito population in the field after low rate (A) and high rate (B) applications of % residual insecticide, calculated using Mulla’s formula: R = 100 ˗ [(C1/T1) × (T2/C2)] × 100; where C1 = pre-treatment measure of mosquito abundance in the associated control site, C2 = post-treatment mosquito abundance in the control site, T1 = pre-treatment mosquito abundance in the treated site, and T2 = post-treatment mosquito abundance in the treated site (Mulla et al. 1971). The corresponding percent reduction is shown with 95% confidence for traps both 6m and 60m away from vegetation treated with OneGuard, Sector, Hyperion, Onslaught Fast Cap, metofluthrin, or a duplex treatment of Sector/Hyperion/Onslaught + metofluthrin. In some instances, population increased in a given week of surveillance, which is displayed in the red region below the 0 on the x-axis.

174

175

Figure A-5. Percent reduction in the relative abundance of Aedes albopictus (Skuse) in the field, as measured by adult (A) and egg (B) % surveillance, calculated using Mulla’s formula: R = 100 ˗ [(C1/T1) × (T2/C2)] × 100; where C1 = pre-treatment measure of mosquito abundance in the associated control site, C2 = post-treatment mosquito abundance in the control site, T1 = pre- treatment mosquito abundance in the treated site, and T2 = post-treatment mosquito abundance in the treated site (Mulla et al. 1971). The corresponding percent reduction is shown with 95% confidence for traps both 6m and 60m away from vegetation treated with a low label rate of either OneGuard, Sector, Hyperion, Onslaught Fast Cap, metofluthrin, or a duplex treatment of Sector/Hyperion/Onslaught + metofluthrin. In some instances, population increased in a given week of surveillance, which is displayed in the red region below the 0 on the x-axis.

176

LIST OF REFERENCES

Abbott, W. S. 1925. A method of computing the effectiveness of an insecticide. J Econ Entomol, 18: 265-267.

Abdel-Mohdy, F. A., M. M. G. Moustafa, M. F. Rehan, and A. S. Aly. 2008. Repellency of controlled-release treated cotton fabrics based on cypermethrin and prallethrin. Carbohydr Polym, 73: 92-97.

Achee, N. L., M. J. Bangs, R. Farlow, G. F. Killeen, S. Lindsay, J. G. Logan, S. J. Moore, M. Rowland, K. Sweeney, S. J. Torr, and J. P. Grieco. 2012a. Spatial repellents: from discovery and development to evidence-based validation. Malaria J, 11: 164. DOI: 10.1186/1475-2875-11-164.

Achee, N. L., P. Masouka, P. Smith, N. Martin, T. Chareonviryiphap, S. Polsomboon, J. Hendarto, and J. P. Grieco. 2012b. Identifying the effective concentration for spatial repellency of the dengue vector Aedes aegypti. Parasit Vectors, 5: 300.

Achee, N. L., M. R. Sardelis, I. Dusfour, K. R. Chauhan, and J. P. Grieco. 2009. Characterization of spatial repellent, contact irritant, and toxicant chemical actions of standard vector control compounds. J Am Mosq Control Assoc, 25: 156.

Adanan, C. R., J. Zairi, and K. H. Ng. 2005. Efficacy and sublethal effects of mosquito mats on Aedes aegypti and Culex quinquefasciatus (Diptera: Culicidae), pp. 265-269. In C. Y. Lee and W. H. Robinson (eds.), Proceedings of the Fifth International Conference on Urban Pests. P & Y Design Network, Perniagaan Ph’ng, Malaysia.

Agramonte, N. M., J. R. Bloomquist, and U. R. Bernier. 2017. Pyrethroid resistance alters the blood-feeding behavior in Puerto Rican Aedes aegypti mosquitoes exposed to treated fabric. PLoS Negl Trop Dis, 11: e0005954. DOI: 10.1371/journal.pntd.0005954

Ahmed, A. M. and H. Hurd. 2006. Immune stimulation and malaria infection impose reproductive costs in Anopheles gambiae via follicular apoptosis. Microbes Infect, 8: 308-315.

Aldridge, R. L., P. E. Kaufman, J. R. Bloomquist, S. A. Gezan, and K. J. Linthicum. 2016. Impact of topical application site on the efficacy of permethrin and malathion to Culex quinquefasciatus. J Am Mosq Control Assoc, 32: 300-307.

Aldridge, R. K., P. E. Kaufman, J. R. Bloomquist, S. A. Gezan, and K. J. Linthicum. 2017. Application site and mosquito age influences malathion- and permethrin-induced mortality in Culex quinquefaciatus (Diptera: Culicidae). J Med Entomol, 54: 1692-1698. DOI: 10.1093/jme/txj160

Allan, S. A., D. L. Kline, and T. Walker. 2009. Environmental factors affecting efficacy of bifenthrin-treated vegetation for mosquito control. J Am Mosq Control Assoc, 25: 338- 347. DOI: 10.2987/09-5854.1 177

Amelia-Yap, Z. H., C. D. Chen, M. Sofian-Azirun, and V. L. Low. 2018. Pyrethroid resistance in the dengue vector Aedes aegypti in Southeast Asia: present situation and prospects for management. Parasite Vector, 11: 332. DOI: 10.1186/s13071-018-2899-0

Andrés, M., L. M. Lorenz, E. Mbeleya, and S. J. Moore. 2015. Modified mosquito landing boxes dispensing transfluthrin provide effective protection against Anopheles arabiensis mosquitoes under simulated outdoor conditions in a semi-field system. Malaria J, 14: 255. DOI: 10.1186/s12936-015-0762-8

Antonio, G. E., D. Sánchez, T. Williams, and C. F. Marina. 2009. Paradoxical effects of sublethal exposure to the naturally derived insecticide spinosad in the dengue vector mosquito, Aedes aegypti. Pest Manag Sci, 65: 323–326.

Argueta, T. B. O., H. Kawada, and M. Takagi. 2004. Spatial repellency of metofluthrin- impregnated multilayer paper against Aedes albopictus under outdoor conditions, Nagasaki, Japan. Med Entomol Zool, 55: 211-216.

Averett, E., J. S. Neuberger, G. Hansen, and M. H. Fox. 2005. Evaluation of a West Nile virus education campaign. Emerg Infect Diseases, 11: 1751-1753. DOI: 10.3201/eid1111.050363

Balabanidou, V., A. Kampouraki, M. MacLean, G. J. Blomquist, C. Tittiger, M. P. Juárez, S. J. Mijailovsky, G. Chalepakis, A. Anthousi, A. Lynd, S. Antoine, J. Hemingway, H. Ranson, G. J. Lycett, and J. Vontas. 2016. Cytochrome P450 associated with insecticide resistance catalyzes cuticular hydrocarbon production in Anopheles gambiae. PNAS, 113: 9268-9273. DOI: 10.1073/pnas.1608295113

Bar-Zeev, M. and D. Ben-Tamar. 1968. The effectiveness of repellents on cloth as determined by oviposition of Aedes aegypti L. Mosq News, 28: 396-403.

Barnes, J. M. and R. D. Verschoyle. 1974. Toxicity of new pyrethroid insecticide. Nature, 248: 711.

Bartlett-Healy, K., G. Hamilton, S. Healy, T. Crepeau, I. Unlu, A. Farajollahi, D. Fonseca, R. Gaugler, G. G. Clark, and D. Strickman. 2011. Source reduction behavior as an independent measurement of the impact of a public health education campaign in an integrated vector management program for the Asian tiger mosquito. Int J Environ Res Pub Heal, 8: 1358–1367. DOI: 10.3390/ijerph8051358

Benedict, M. Q., R. S. Levine, W. A. Hawley, and P. Lounibos. 2007. Spread of the tiger: global risk of invasion by the mosquito Aedes albopictus. Vector Borne Zoonotic Dis, 7: 76-85. DOI: 10.1089/vbz.2006.0562

Bibbs C. S., C. Anderson, M. Smith and R. D. Xue. 2016. Bifenthrin barrier spray against Aedes albopictus (Skuse) around an urban cemetery, St. Augustine, FL. Technical Bulletin for the Florida Mosquito Control Association, 10: 76-80.

178

Bibbs, C. S., J. R. Bloomquist, D. A. Hahn, P. E. Kaufman, and R. D. Xue. 2019a. Gone in 60 seconds: sub-lethal effects of metofluthrin vapors on behavior and fitness of resistant and field strains of Aedes aegypti (Diptera: Culicidae). J Med Entomol, 56: 1087-1094. DOI: 10.1093/jme/tjz040.

Bibbs, C. S., A. P. Fulcher, and R. D. Xue. 2015. Allethrin based mosquito control device causing knockdown, morbidity, and mortality in four species of field-caught mosquitoes (Diptera: Culicidae). J Med Entomol 52: 739-742. DOI: 10.1093/jme/tjv065.

Bibbs C. S., D. A. Hahn, P. E. Kaufman, and R. D. Xue. 2018a. Sub-lethal effects of a vapor- active pyrethroid on Aedes aegypti and Ae. albopictus fecundity and oviposition behavior (Diptera: Culicidae). Parasit Vectors, 11: 486. DOI: http://dx.doi.org/10.1186/s13071- 018-3065-4.

Bibbs C. S. and P. E. Kaufman. 2017. Volatile pyrethroids as a potential mosquito abatement tool: a review of pyrethroid containing spatial repellents. J Integr Pest Manag, 8: 21; 1- 10. DOI: http://dx.doi.org/10.1093/jipm/pmx016

Bibbs, C. S., J. D. Kline, D. L. Kline, J. Estaver, R. Strohschein, S. A. Allan, P. E. Kaufman, R. D. Xue, and C. D. Batich. 2019b. Olfactometric comparison of the volatile insecticide, metofluthrin, through behavioral responses of Aedes albopictus (Diptera: Culicidae). Journal of Medical Entomology, 56: in press. DOI: 10.1093/jme/tjz092.

Bibbs C. S., M. Tsikolia, U. R. Bernier, J. R. Bloomquist, R. D. Xue, P. E. Kaufman. 2018b. Vapor toxicity of five volatile pyrethroids against Aedes aegypti, Ae. albopictus, Culex quinquefasciatus, and Anopheles quadrimaculatus (Diptera: Culicidae). Pest Manag Sci, 74: 2699–2706. DOI: 10.1002/ps.5088.

Bibbs, C. S. and R. D. Xue. 2015. OFF! Clip-on repellant device with metofluthrin tested on Aedes aegypti (Diptera: Culicidae) for mortality at different time intervals and distances. J Med Entomol, 52: 480-483. DOI: 10.1093/jme/tjv200

Bibbs, C. S. and R. D. Xue. 2016. ThermaCELL and OFF! clip-on devices tested for repellency and mortality against Amblyomma americanum (Ixodida: Amblyommidae). J Med Entomol, 53: 861-865. DOI: 10.1093/jme/tjw039

Bodner, D., S. L. LaDeau, D. Biehler, N. Kirchoff, and P. T. Leisnham. 2016. Effectiveness of print education at reducing urban mosquito infestation through improved resident- based management. PLoS ONE, 11: e0155011. DOI: 10.1371/journal.pone.0155011

Bong, L. J., W. C. Tu, K. B. Neoh, C. G. Huang, and R. X. Ting. 2017. The effect of insecticidal stress on reproductive output of susceptible and field strains of Aedes aegypti (Diptera: Culicidae). J Med Entomol, 55: 36-42. DOI: 10.1093/jme/tjx191

179

Bowman, N. M., K. Akialis, G. Cave, R. Barrera, C. S. Apperson, and S. R. Meshnick. 2018. Pyrethroid insecticides maintain repellent effect on knock-down resistant populations of Aedes aegypti mosquitoes. PLoS ONE, 13: e0196410. DOI: 10.1371/journal.pone.0196410

Britch, S. C., K. J. Linthicum, R. L. Aldridge, M. S. Breidenbaugh, M. D. Latham, P. H. Connelly, M. J. E. Rush, J. L. Remmers, J. D. Kerce, and C. A. Silcox. 2018. Aerial ULV control of Aedes aegypti with naled (Dibrom) inside simulated rural village and urban cryptic habitats. PLoS ONE, 13: e0191555.

Buhagiar, T. S., G. J. Devine, and S. A. Ritchie. 2017a. Metofluthrin: investigations into the use of a volatile spatial pyrethroid in a global spread of dengue, chikungunya and Zika viruses. Parasite Vector, 10: 270. DOI: 10.1186/s13071-017-2219-0.

Buhagiar, T. S., G. J. Devine, and S. A. Ritchie. 2017b. Effects of sublethal exposure to metofluthrin on the fitness of Aedes aegypti in a domestic setting in Cairns, Queensland. Parasit Vectors, 10: 274. DOI: 10.1186/s13071-017-2220-7.

(CDC) United States Centers for Disease Control and Prevention. 2003. Local transmission of Plasmodium vivax Malaria --- Palm Beach County, Florida, 2003. MMWR, 52 (38): 908-911.

(CDC) United States Centers for Disease Control and Prevention. 2016. Responding to local mosquito-borne transmission of Zika virus, pp. 1-22. In Operational Risk Communication and Community Engagement Plan. United States Department of Health and Human Services, Atlanta, GA.

(CDC) United States Centers for Disease Control and Prevention. 2018. Tick bite prevention and prophylaxis, pp. 48-49. In Tickborne diseases of the US: a reference manual for health care providers, fifth edition. United States Department of Health and Human Services, Atlanta, GA.

Chareonviriyaphap, T. 2012. Behavioral responses of mosquitoes to insecticides, pp. 201-220. In F. Perveen (ed.), Insecticides: Pest Engineering. InTech, Rijeka, Croatia.

Chareonviriyaphap, T., B. Aum-Aong, and S. Ratanatham. 1999. Current insecticide resistant pattern in mosquito vectors. Southeast Asian J Trop Med Public Health, 30: 130- 141.

Chareonviriyaphap, T., S. Sungvornyothin, S. Ratanatham, and A. Prabaripai. 2001. Pesticide-induced behavioral responses of Anopheles minimus, a malaria vector in Thailand. J Am Mosq Control Assoc, 17: 13-22.

Chen, Y. S. 1984. Metabolism and breakdown products of Dibrom 14. J Florida Anti-Mosq Assoc, 55: 46-47.

180

Choi, D. B., J. P. Grieco, C. S. Apperson, C. Schal, L. Ponnusamy, D. M. Wesson, and N. L. Achee. 2016. Effect of spatial repellent exposure on dengue vector attraction to oviposition sites. PLoS Negl Trop Dis, 10: e0004850. DOI: 10.1371/journal.pntd.0004850.

Chouin-Carneiro, T., A. Vega-Rua, M. Vazeille, A. Yebakima, R. Girod, D. Goindin, M. Dupont-Rouzeyrol, R. Lourenco-de-Oliveira, and A. B. Failloux. 2016. Differential susceptibilities of Aedes aegypti and Aedes albopictus from the Americas to Zika virus. PLoS Negl Trop Dis, 11: 0004543

Christophers, S. R. 1947. Mosquito repellents. Being a report of the work of the mosquito repellent inquiry, Cambridge 1943-5. J Hyg, 45: 176.

Cilek, J. E. 2008. Application of insecticides to vegetation as barriers against host-seeking mosquitoes. J Am Mosq Control Assoc, 24: 172-176.

Cilek, J. E. and C. F. Hallmon. 2008. Residual effectiveness of three pyrethroids on vegetation against adult Aedes albopictus and Culex quinquefasciatus in screened field cages. J Am Mosq Control Assoc, 24: 263-269.

Cohnstaedt, L. W. and S. A. Allan. 2011. Effects of sublethal pyrethroid exposure on the host- seeking behavior of female mosquitoes. J Vec Ecol, 36: 395-403.

Cook, S., Z. R. Khan, and J. A. Pickett. 2007. The use of push-pull strategies in integrated pest management. Ann Rev Entomol, 52: 375-400.

Cutler, G. C. 2013. Insects, insecticides and hormesis: evidence and considerations for study. Dose Response, 11: 154-177. DOI: 10.2203/dose-response.12-008.Cutler

Dame, D. A., M. V. Meisch, C. N. Lewis, D. L. Kline, and G. G. Clark. 2014. Field evaluation of four spatial repellent devices against Arkansas rice-land mosquitoes. J Am Mosq Control Assoc, 30: 31-36.

Davis, T. J., P. E. Kaufman, A. J. Tatem, J. A. Hogsette, and D. L. Kline. 2016. Development and evaluation of an attractive self-marking ovitrap to measure dispersal and determine skip oviposition in Aedes albopictus (Diptera: Cluicidae) field populations. J Med Entomol, 53: 31-38.

Day, J. F. and L. M. Stark. 1999. Avian serology in a St. Louis encephalitis epicenter before, during, and after a widespread epidemic in South Florida, USA. J Med Entomol, 36: 614– 24.

Deguchi, Y., T. Yamada, Y. Hirose, H. Nagahori, M. Kushida, K. Sumida, T. Sukata, Y. Tomigahara, K. Nishioka, S. Uwagawa, S. Kawamura, and Y. Okuno. 2009. Mode of action anlaysis for the synthetic pyrethroid metofluthrin-induced rat liver tumors: evidence for hepatic CYP2B induction and hepatocyte proliferation. Toxicol Sci, 108: 69-80. DOI: 10.1093/toxsci/kfp006 181

Dekel, A., R. J. Pitts, E. Yakir, and J. D. Bohbot. 2016. Evolutionarily conserved odorant receptor function questions ecological context of octanol role in mosquitoes. Scientific Reports, 6: 37330.

Derraik, J. G. and D. Slaney. 2015. Notes on Zika virus—an emerging pathogen in the South Pacific Aust N Z J Public Health 39: 5-7.

Devine, G. J., E. Z. Perea, G. F. Killeen, J. D. Stancil, S. J. Clark, and A. C. Morrison. 2009. Using adult mosquitoes to transfer insecticides to Aedes aegypti larval habitats. Proc Natl Acad Sci USA, 106: 11530–11534.

Dewer, Y., M. A. Pottier, L. Lalouette, A. Maria, M. Dacher, L. P. Belzunces, G. Kairo, D. Renault, M. Maibeche, and D. Siaussat. 2016. Behavioral and metabolic effects of sublethal doses of two insecticides, chlorpyrifos and methomyl, in the Egyptian cotton leafworm, Spodoptera littoralis (Boisduval) (Lepidopters: Noctuidae). Environ Sci Pollut Res, 23: 3086-3096. DOI: 10.1007/s11356-015-5710-1.

Dye-Braumuller, K. C., K. F. Jaynes, and G. C. Brown. 2017. Quantitative analysis of Aedes albopictus movement behavior following sublethal exposure to prallethrin. J Am Mosq Control Assoc, 33: 282-292.

Edman, J. D., E. Cody, and H. Lynn. 1975. Blood-feeding activity of partially engorged Culex nigripalpus (Diptera: Culicidae). Entomol Exp Appl, 18: 261-268.

Elliot, M., A. W. Farnham, N. F. Janes, P. H. Needham, and D. A. Pulman. 1973. Potent pyrethroid insecticides from modified cyclopropane acids. Nature, 144: 456-457.

(EPA) United States Environmental Protection Agency. 2006. Pesticide Fact Sheet: 7505P, Metofluthrin. Office of Prevention, Pesticides, and Toxic Substances, United States Environmental Protection Agency, Washington, D.C.

(EPA) United States Environmental Protection Agency. 2009. Product performance test guidelines: OPPTS 810.3400 mosquito, black fly, and biting midge (sand fly) treatments. Office of Prevention, Pesticides, and Toxic Substances, United States Environmental Protection Agency, Washington, D.C.

Estaver, J. and R. Strohschein. 2019. Sigma Scientific LLC product catalogue for insect- stimulus response systems, Revision 3. https://www.sigmascientificllc.com/, Last accessed Feb 11, 2019.

Estep, A. S., N. D. Sanscrainte, C. M. Waits, S. J. Bernard, A. M. Lloyd, K. J. Lucas, E. A. Buckner, R. Vaidyanathan, R. Morreale, L. A. Conti, and J. J. Becnel. 2018. Quantification of permethrin resistance and kdr alleles in Florida strains of Aedes aegypti (L.) and Aedes albopictus (Skuse). PLoS Neg Trop Dis, 12: e0006544. DOI: 10.1371/journal.pntd.0006544

182

Estep, A. S., N. D. Sanscrainte, C. M. Waits, J. E. Louton, and J. J. Becnel. 2017. Resistance status and resistance mechanisms in a strain of Aedes aegypti (Diptera: Culicidae) from Puerto Rico. J Med Entomol, 54: 1643-1648. DOI: 10.1093/jme/tjx143

Faraji, A. and I. Unlu. 2016. The eye of the tiger, the thrill of the fight: effective larval and adult control measures against the Asian tiger mosquito, Aedes albopictus (Diptera: Culicidae), in North America. J Med Entomol, 53: 1029-1047. DOI: 10.1093/jme/tjw096

Faraji, A., I. Unlu, T. Crepeau, S. Healy, S. Crans, G. Lizarraga, D. Fonseca, and R. Gaugler. 2016. Droplet characterization and penetration of an ultra-low volume mosquito adulticide spray targeting the Asian tiger mosquito, Aedes albopictus, within urban and suburban environments of northeastern USA. PLoS ONE, 11: e0152069

Farajollahi (Faraji) A., S. P. Healy, I. Unlu, R. Gaugler, and D. M. Fonseca. 2012. Eﬀectiveness of ultra-low volume nighttime applications of an adulticide against diurnal Aedes albopictus, a critical vector of dengue and chikungunya viruses. PLoS ONE, 7:e49181.

Fay, R. W. and A. S. Perry. 1965. Laboratory studies of oviposition preferences of Aedes aegypti. Mosq News, 25: 276-281.

Fonseca, D. M., L. R. Kaplan, R. A. Heiry, and D. Strickman. 2015. Density-dependent oviposition by female Aedes albopictus (Diptera: Culicidae) spreads eggs among containers during the summer but accumulates them in the fall. J Med Entomol, 52: 705- 712.

Fourie, J. J., I. G. Horak, C. de Vos, K. Desuter, and B. Schunack. 2015. Comparative speed of kill, repellent (anti-feeding) and acaricidal efficacy of an imidacloprid/flumethrin collar (Seresto®) and a fipronil/(s)-methoprene/eprinomectin/ praziquantel spot-on (Broadline®) against Ixodes ricinus (Linné, 1758) on cats. Parasitol Res 114: S109-S116. DOI 10.1007/s00436-015-4517-9.

Gibson, J., R. D. Xue, and M. L. Smith. 2016. Field comparison of thermal fog application of Duet® and barrier spraying of Talstar® against Aedes albopictus at a residential property in St. Augustine, FL. Technical Bulletin for the Florida Mosquito Control Association, 10: 96-98.

Grieco, J. P., N. L. Achee, T. Chareonviriyaphap, W. Suwonkerd, K. Chauhan, M. R. Sardelis, and D. R. Roberts. 2007. A new classification system for the actions of IRS chemicals traditionally used for malaria control. PLoS One, 2: e716.

Gross, A. D. and J. R. Coats. 2015. Can green chemistry provide effective repellents? Insect Repellents Handbook (ed. by M. Debboun, S.P. Frances & D.A. Strickman), pp. 75-90. CRC Press and Taylor and Francis Group LLC, Boca Raton, FL

183

Gross, A. D., E. J. Norris, M. J. Kimber, L. C. Bartholomay, and J. R. Coats. 2017. Essential oils enhance the toxicity of permethrin against Aedes aegypti and Anopheles gambiae. Med Vet Entomol, 34: 55-62.

Guo, J. 2016. Studies using IPS cells support a possible link between ZIKA and microcephaly. Cell Biosci, 6: 28. DOI: 10.1186/s13578-016-0096-4.

Hall, G. L., D. Fitzell, C. R. Mourer, and T. Shibamoto. 1997. Development and validation of an analytical method for naled and dichlorvos in air. J Agric Food Chemi, 45: 145-148.

Hao, H., J. Wei, J. Dai, and J. Du. 2008. Host-seeking and blood-feeding behavior of Aedes albopictus (Diptera: Culicidae) exposed to vapors of geraniol, citral, citronella, eugenol, or anisaldehyde. J Med Entomol, 45: 533-539.

Harrington, L. C. and J. D. Edman. 2001. Indirect evidence against delayed “skip-oviposition” behavior by Aedes aegypti (Diptera: Culicidae) in Thailand. J Med Entomol, 38: 641-645.

Harrison, C. M. 1951. Inheritance of resistance of DDT in the housefly, Musca domestica L. Nature, 167 (4256): 855-856.

Hill, C. A., F. C. Kafatos, S. K. Stansfield, and F. H. Collins. 2005. Arthropod-borne disease: vector control in the genomics era. Nat Rev Microbiol, 3: 262-268.

Hirose, Y., H. Nagahori, T. Yamada, Y. Deguchi, Y. Tomigahara, K. Nishioka, S. Uwagawa, S. Kawamura, N. Isobe, B. G. Lake, and Y. Okuno. 2009. Comparison of the effects of the synthetic pyrethroid metofluthrin and phenobarbital on CYP2B form induction and replicative DNA synthesis in cultured rat and human hepatocytes. Toxicology, 258: 64-69. DOI: 10.1016/j.tox.2009.01.007

Holderman, C. J., S. A. Gezan, A. E. S. Stone, C. R. Connelly, and P. E. Kaufman. 2018. Mosquitoes (Diptera: Culicidae) collected from residential yards and dog kennels in Florida using two aspirators, a sweep net, or a CDC trap. J Med Entomol, 55: 230-236. DOI: 10.1093/jme/tjx171

Honório, N. A., W. da Costa Silva, P. J. Leite, J. M. Gonçalves, L. P. Lounibos, and R. Lourenço-de-Oliveira. 2003. Dispersal of Aedes aegypti and Aedes albopictus (Diptera: Culicidae) in an urban endemic dengue area in the state of Rio de Janeiro, Brazil. Mem Inst Oswaldo Cruz Rio de Janeiro, 98: 191-198.

Horstmann S. and R. Sonneck 2016. Contact bioassays with phenoxybenzyl and tetrafluorobenzyl pyrethroids against target-site and metabolic resistant mosquitoes. PLoS ONE 11(3): e0149738. DOI: 10.1371/journal.pone.0149738

Huho, B. P., H. Ngonyani, E. Simfukwe, E. Mbeyela, and S. Moore. 2015. Protective efficacy of a transfluthrin-based spatial repellent in combination with long lasting insecticide treated nets. Trop Med International Health, 20: 402-402.

184

Jalilnavaz, M. R., M. R. Abai, H. Vatandoost, M. Mohebali, A. A. Akhavan, Z. Zarei, S. Rafizadeh, H. Bakhshi, and Y. Rassi. 2016. Application of flumethrin pour-on on reservoir dogs and its efficacy against sand flies in endemic focus of visceral leishmaniasis, Meshkinshahr, Iran. J Arthropo-Borne Dis, 10: 78-86.

Jiang, S., L. Yang, and J. R. Bloomquist. 2019. High-throughput screening method for evaluating spatial repellency and vapor toxicity to mosquitoes. Med Vet Entomol, 33: DOI: 10.1111/mve.12377

Katsuda, Y, S. Leemingsawat, S. Thongrunkiat, S. Prummonkol, Y. Samnug, T. Kanzaki, T. Watanabe. 2009. Control of mosquito vectors of tropical infectious diseases: (3) susceptibility of Aedes aegypti to pyrethroid and mosquito coils. Southeast Asian J Trop Med Public Health 40: 929-936.

Kawada, H., T. Iwasaki, L. L. Loan, T. K. Tien, N. T. N. Mai, Y. Shono, Y. Katayama, and M. Takagi. 2006. Field evaluation of spatial repellency of metofluthrin-impregnated latticework plastic strips against Aedes aegypti (L.) and analysis of environmental factors affecting its efficacy in My Tho City, Tien Giang, Vietnam. Am J Trop Med Hyg 75: 1153–1157.

Kawada, H., Y. Maekawa, and M. Takagi. 2005. Field trial on the spatial repellency of metofluthrin-impregnated plastic strips for mosquitoes in shelters without walls (beruga) in Lombok, Indonesia. J Vector Ecol, 30: 181-185.

Kawada, H., E. A. Temu, J. N. Minjas, O. Matsumoto, T. Iwasaki, and M. Takagi. 2008. Field evaluation of spatial repellency of metofluthrin-impregnated plastic strips against Anopheles gambiae complex in Bagamoyo, coastal Tanzania. J Am Mosq Control Assoc, 24: 404-409.

Kelly, D. W. and C. E. Thompson. 2000. Epidemiology and optimal foraging: modelling the ideal free distribution of insect vectors. Parasitology 120: 319–327.

Kline, D. L., U. R. Bernier, K. H. Posey, and D. R. Barnard. 2003. Olfactometric evaluation of spatial repellents for Aedes aegypti. J Med Entomol, 40: 463-467.

Klowden, M. J. and A. O. Lea. 1978. Blood meal size as a factor affecting continued host seeking by Aedes aegypti (L.). Am J Trop Med Hyg, 27: 827-831

Lee, D. K. 2007. Lethal and repellent effects of transfluthrin and metofluthrin used in portable blowers for personal protection against Ochlerotatus togoi and Aedes albopictus (Diptera: Culicidae). Entomol Res, 37: 173-179.

Leontsini, E., E. Gil, and G. G. Clark. 1993. Effect of a community-based Aedes aegypti control program on mosquito larval production in El Progreso Honduras. Trans Roy Soc Trop Med Hyg, 87: 267–271. DOI: 10.1016/0035-9203(93)90120-f

185

Liu, W. and G. P. Georghiou. 1987. Influence of permethrin treatment on blood feeding and fecundity of Culex quinquefasciatus Say resistant and susceptible to pyrethroid insecticides. Contrib Shanghai Inst of Entomol, 7: 101-109

Liu, W., R. G. Todd, and E. J. Gerberg. 1986. Effect of three pyrethroids on blood feeding and fecundity of Aedes aegypti. J Am Mosq Control Assoc, 2: 310-313.

Liu, W. H., S. W. Cai, J. Wu, L. F. Lin, J. H. Duan, and Q. Zou. 2009. Research on the efficacy of four kinds of electric liquid repellents for the Japanese encephalitis vectors in the field. Chinese J Vector Bio Control, 20: 505-506.

Lucas, J. R., Y. Shono, T. Iwasaki, T. Ishiwatari, N. Spero, and G. Benzon. 2007. US laboratory and field trials of metofluthrin (SumiOne®) emanators for reducing mosquito biting outdoors. J Am Mosq Control Assoc, 23: 47-54.

Lumsden, L. L. 1958. St. Louis encephalitis in 1933: observations in epidemiological features. Pub Health Reports, 73: 340-353.

Maciel-de-Freitas, R. and R. Lourenço-de-Oliveira. 2011. Does targeting key-containers effectively reduce Aedes aegypti population density? Trop Med Int Health, 16: 965-973.

Maia, M. F., K. Kreppel, E. Mbeyela, D. Roman, V. Mayagaya, N. F. Lobo, A. Ross, and S. J. Moore. 2016. A crossover study to evaluation the diversion of malaria vectors in a community with incomplete coverage of spatial repellents in Kilombera Valley, Tanzania. Parasites Vectors, 9: 451. DOI: 10.1186/s13071-016-1738-4

Manda, H., P. Shah, S. Polsomboon, T. Chareonviriyaphap, F. Castro-Llanos, A. Morrison, R. G. Burrus, J. P. Grieco, and N. L. Achee. 2013. Contact irritant responses of Aedes aegypti using sub-lethal concentration and focal application of pyrethroid chemicals. PLoS Negl Trop Dis, 7:e2074. DOI: 10.1371/journal.pntd.0002074 PMID: 23469302

Menger, D. J. 2015. Push-pull tactics to disrupt the host-seeking behavior of malaria mosquitoes. Wageningen University, Wageningen, NL.

Miyamoto, J. 1976. Degradation, metabolism, and toxicity of synthetic pyrethroids. Environ Health Persp, 14: 15-28.

Moore, S. J., C. R. Davies, N. Hill, and M. M. Cameron. 2007. Are mosquitoes diverted from repellent-using individuals to non-users? Results of a field study in Bolivia. Trop Med Int Health, 12: 532-539. DOI: 10.1111/j.1365-3156.2006.01811.x

Msangi, S., B. Mwang’onde, A. M. Mahande, and E. J. Kweka. 2010. Field evaluation of the bio-efficacy of three pyrethroid based coils against wild populations of anthropophilic mosquitoes in northern Tanzania. J Glob Infect Dis, 2: 116-120.

Mulla M. S., R. L. Norland, D. M. Fanara, H. A. Darwezeh, and D. W. McKean. 1971. Control of chironomid midges in recreational lakes. J Econ Entomol, 64: 300–307.

186

Mwandawiro, C., M. Boots, N. Tuno, W. Suwonkerd, Y. Tsunda, and M. Takagi. 2000. Heterogeneity in the host preferences of Japanese encephalitis vectors in Chiang Mai, northern Thailand. Trans R Soc Trop Med Hyg 94: 238–242.

Naumann, K. 1998. Research into fluorinated pyrethroid alcohols - an episode in the history of pyrethroid discovery. Pest Sci, 52: 3-20.

Ngaogouni, C., B. Kamgang, E. Nakoune, C. Paupy, M. Kazanji. 2015. Invasion of Aedes albopictus (Diptera: Culicidae) into central Africa: what consequences for emerging diseases? Parasit Vectors, 8: 191.

Nikbakhtzadeh, M. R., J. W. Terbot II, P. E. Otienoburu, and W. A. Foster. 2014. Olfactory basis of floral preference of the malaria vector Anopheles gambiae (Diptera: Culicidae) among common African plants. J Vector Ecol, 39: 372-383.

Noori, N., B. G. Lockaby, and L. Kalin. 2015. Larval development of Culex quinquefasciatus in water with low to moderate pollution levels. J Vector Ecol, 40: 208-220.

Norris, E. J., A. D. Gross, B. Dunphy, L. C. Bartholomay, and J. R. Coats. 2015. Comparison of the insecticidal characteristics of commercially available plant essential oils against Aedes aegypti and Anopheles gambiae. J Med Entomol, 52: 993-1002.

Norris, E. J., J. B. Johnson, A. D. Gross, L. C. Bartholomay, and J. R. Coats. 2018. Plant essential oils enhance diverse pyrethroids against multiple strains of mosquitoes and inhibit detoxification enzyme processes. Insects, 9: 132. DOI: 10.3390/insects9040132

Obermayr, U., J. Ruther, U. R. Bernier, A. Rose, M. Geier. 2015. Evaluation of a push-pull approach for Aedes aegypti (L.) using a novel dispensing system for spatial repellents in the laboratory and in a semi-field environment. PLoS ONE, 10: e0129878. DOI: 10.1371/journal.pone.0129878.

Offerdahl, D. K., D. W. Dorward, B. T. Hansen, and M. E. Bloom. 2016. Cytoarchitecture of Zika virus infection in human neuroblastoma and Aedes albopictus cell lines. Virology, 501: 54-62.

Ogoma, S. B., L. M. Lorenz, H. Ngonyani, R. Sangusangu, M. Kitumbukile, M. Kilalangongono, E. T. Simfukwe, A. Mseka, E. Mbeyela, D. Roman, J. Moore, K. Kreppel, M. F. Maia, and S. J. Moore. 2014a. An experimental hut study to quantify the effect of DDT and airborne pyrethroids on entomological parameters of malaria transmission. Malar J, 13: 131.

Ogoma, S. B., A. S. Mmando, J. K. Swai, S. Horstmann, D. Malone, G. F. Killeen. 2017. A low technology emanator treated with the volatile pyrethroid transfluthrin confers long term protection against outdoor biting vectors of lymphatic filariasis, arboviruses and malaria. PLoS Negl Trop Dis, 11: e0005455. DOI: 10.1371/journal.pntd.0005455.

187

Ogoma, S. B., H. Ngonyani, E. T. Simfukwe, A. Mseka, J. Moore, G. F. Killeen. 2012. Spatial repellency of transfluthrin-treated hessian strips against laboratory-reared Anopheles arabiensis mosquitoes in a semi-field tunnel cage. Parasit Vectors, 5: 54.

Ogoma, S. B., H. Ngonyani, E. Simfukwe, A. Mseka, J. Moore, M. F. Maia, S. J. Moore, and L. M. Lorenz. 2014b. The mode of action of spatial repellents and their impact on vectorial capacity of Anopheles gambiae sensu stricto. PLoS ONE, 9: e110433. DOI: 10.1371/journal.pone.0110433

Ohara, A., Y. Takahashi, M. Kondo, Y. Okuda, S. Takeda, M. Kushida, K. Kobayashi, K. Sumida, and T. Yamada. 2017. Candidate genes responsible for early key events of phenobarbital-promoted mouse hepatocellular tumorigenesis based on differentiation of regulating genes between wild type mice and humanized chimeric mice. Toxicol Research, 6: 795. DOI: 10.1039/c7tx00163k

Okuda, Y., M. Kushida, H. Kikumoto, Y. Nakamura, H. Higuchi, S. Kawamura, S. M. Cohen, B. G. Lake, and T. Yamada. 2017. Evaluation of the human relevance of the constitutive androstane receptor-mediated mode of action for rat hepatocellular tumor formation by the synthetic pyrethroid momfluorothrin. J Toxicol Sci, 42: 773-788.

Oliosi, E., C. S. Corcos, P. F. Barroso, A. Bleibtreu, G. Grard, B. A. M. De Filippis, and E. Caumes. 2018. Yellow fever in two unvaccinated French tourists to Brazil, January and March, 2018. Eurosurveillance, 23: 2-5.

Oliva, L. O., J. C. Correia, and C. M. R. Albuquerque. 2014. How does mosquito age and the type and color of oviposition sites modify skip-oviposition behavior in Aedes aegypti (Diptera: Culicidae)? J Insect Behav, 27: 81-91.

Paliga, J. M., M. F. Maia, M. Sikulu, F. O. Okumu, and S. B. Ogoma. 2015. The complementary advantage of combining spatial repellent treated sisal decorative baskets with long lasting insecticide treated nets. Trop Med International Health, 20: 110.

Pates, H. V., J. D. Lines, A. J. Keto, and J. E. Miller. 2002. Personal protection against mosquitoes in Dar es Salaam, Tanzania, by using a kerosene oil lamp to vaporize transfluthrin. Med Vet Entomol, 16: 277–84.

Paz-Soldan, V. A., V. Plasai, A. C. Morrison, E. J. Rios-Lopez, S. Guedez-Gonzales, J. P. Grieco, K. Mundal, T. Chareonviriyaphap, and N. L. Achee. 2011. Initial assessment of the acceptability of a push-pull Aedes aegypti control strategy in Iquitos, Peru and Kanchanaburi, Thailand. Am J Trop Med Hyg, 84: 208-217. DOI: 10.4269/ajtmh.2011.09-0615

Ponlawat, A., P. Kankaew, S. Chanaimongkol, A. Pongsiri, J. H. Richardson, and B. P. Evans. 2016. Semi-field evaluation of metofluthrin-impregnated nets on host-seeking Aedes aegypti and Anopheles dirus. J Am Mosq Control Assoc, 32: 130-138. DOI: https://doi.org/10.2987/moco-32-02-130-138.1

188

Posey K. H. and C. E. Schreck. 1981. An airflow apparatus for selecting female mosquitoes for use in repellent and attraction studies. Mosq News, 41: 566–568.

Rapley, L. P., R. C. Russell, B. L. Montgomery, S. A. Ritchie. 2009. The effects of sustained release metofluthrin on the biting, movement, and mortality of Aedes aegypti in a domestic setting. Am J Trop Med Hyg, 81: 94-99.

Rawlinson, G. 1996. The histories of Herodotus, 440 B.C. Penguin, London, UK, reprint.

Revay, E. E., D. L. Kline, R. D. Xue, W. A. Qualls, U. R. Bernier, V. D. Kravchenko, N. Ghattas, I. Pstygo, and G. C. Müller. 2013. Reduction of mosquito biting-pressure: Spatial repellents or mosquito traps? A field comparison of seven commercially available products in Israel. Acta Tropica, 127: 63-68.

Rey, J. R. 2014. Dengue in Florida (USA). Insects, 5: 991-1000.

Ritchie, S. A. and G. J. Devine. 2013. Confusion, knock-down and kill of Aedes aegypti using metofluthrin in domestic settings: a powerful tool to prevent dengue transmission? Parasite Vector, 6: 262.

Roberts, D. R., W. D. Alecrim, A. M. Tavares, and K. M. McNeil. 1984. Influence of physiological condition on the behavioral responses of Anopheles darlingi to DDT. Mosq News, 44: 357-361.

Roberts, D. R., W. D. Alecrim, P. Hshieh, J. Grieco, M. J. Bangs, R. G. Andre, and T. Chareonviriyaphap. 2000. A probability model of vector behavior: effects of DDT repellency, irritability, and toxicity in malaria control. J Vec Ecol, 25: 48-61.

Rutledge, R. C., R. H. Baker, C. D. Morris, and J. K. Nayar. 2005. Human malaria. University of Florida, EDIS, MG103. http://edis.ifas.ufl.edu/MG103 (Last accessed 28 Jan 2019).

Salazar, F. V., N. L. Achee, J. P. Grieco, A. Prabaripai, T. A. Ojo, L. Eisen, C. Dureza, S. Polsomboon, and T. Chareonviriyaphap. 2013. Effect of Aedes aegypti exposure to spatial repellent chemicals on BG-Sentinel trap catches. Parasite Vector, 6: 145.

Santos de Abreu, F. V., M. M. Morais, S. P. Ribeiro, and A. E. Eiras. 2015. Influence of breeding site availability on the oviposition behavior of Aedes aegypti. Mem Inst Oswaldo Cruz Rio de Janeiro, 110: 669-676.

Sathantriphop, S., S. A. White, N. L. Achee, U. Sanguanpong, and T. Chareonviriyaphap. 2014. Behavioral responses of Aedes aegypti, Aedes albopictus, Culex quinquefasciatus, and Anopheles minimus against various synthetic and natural repellent compounds. J Vec Ecol, 39: 328-339.

189

Shen Y., R. D. Xue, and C. S. Bibbs. 2017. Relative insecticidal efficacy of three spatial repellent integrated light sources against Aedes aegypti (Diptera: Culicidae). J Am Mosq Control Assoc, 33: 348-351.

Siria, D. J., E. P. A. Batista, M. A. Opiyo, E. F. Melo, R. D. Sumaye, H. S. Ngowo, A. E. Eiras, and F. O. Okumu. 2018. Evaluation of a simple polytetrafluoroethylene (PTFE)- based membrane for blood-feeding of malaria and dengue fever vectors in the laboratory. Parasite Vector, 11: 236. DOI: 10.1186/s13071-018-2823-7

Smith, T. J. and D. M. Soderlund. 1998. Actions of the pyrethroid insecticides cismethrin and cypermethrin on house fly Vssc1 sodium channels expressed in Xenopus oocytes. Arch Insect Biochem Physiol, 38: 126-136.

Smith M. L., D. Dixon, C. S. Bibbs, D. Autry, and R. D. Xue. 2018. Diel activity patterns of Aedes aegypti after resurgence in St. Augustine, Florida as collected by a mechanical rotator trap. J Vector Ecol, 43: 200-204.

Sparks, T. C. 2013. Insecticide discovery: an evaluation and analysis. Pest Biochem Physiol, 107: 8-17.

Stafford, K. C. 2007. Area-wide chemical control of ticks, pp. 63-70. In Tick management handbook. Connecticut Agricultural Experiment Station, Connecticut General Assembly, New Haven, CT.

Stanczyk, N. M., J. F. Brookfield, L. M. Field, and J. G. Logan. 2013. Aedes aegypti mosquitoes exhibit decreased repellency by DEET following previous exposure. PLoS ONE, 8: e54438.

Stevenson, J. C., L. Simubali, T. Mudenda, E. Cardol, U. R. Bernier, A. A. Vasquez, P. E. Thuma, D. E. Norris, M. Perry, D. L. Kline, L. W. Cohnstaedt, P. Gurman, S. D’hers, and N. M. Elman. 2018. Controlled release spatial repellent devices (CRDs) as novel tools against malaria transmission: a semi-field study in Macha, Zambia. Malara J, 17: 437. DOI: 10.1186/s12936-018-2558-0

Subra, R. 1981. Biology and control of Culex pipens quinquefasciatus Say, 1823 (Diptera, Culicidae) with special reference to Africa. Insect Sci Application, 1: 319-338.

Sugano, M. and T. Ishiwatari. 2012. The biological activity of a novel pyrethroid: metofluthrin. Top Curr Chem, 314: 203-220. DOI: 10.1007/128_2011_259

Sugiharto, V. A., J. P. Greico, J. R. Murphy, C. H. Olsen, M. G. Colacicco-Mayhugh, V. A. Stewart, N. L. Achee, and M. J. Turell. 2016. Effects of preexposure to DEET on the downstream blood-feeding behaviors of Aedes aegypti (Diptera: Culicidae) mosquitoes. J Med Entomol, 53: 1100-1104. DOI: 10.1093/jme/tjw066

190

Suman, D. S., A. Farajollahi, S. Healy, G. M. Williams, Y. Wang, G. Schoeler, and R. Gaugler. 2014. Point-source and area-wide field studies of pyriproxyfen autodissemination against urban container-inhabiting mosquitoes. Acta Tropica, 135: 96- 103.

Sumita, Y., H. Kawada, and N. Minakawa. 2016. Mode of entry of vaporized pyrethroid knockdown agent into the body of the housefly, Musca domestica (Diptera: Muscidae). Appl Entomol Zool, 51: 653-659.

Syafruddin, D., M. J. Bangs, D. Sidik, I. Elyazar, P. B. S. Asih, K. Chan, S. Nurleila, C. Nixon, J. Hendarto, I. Wahid, H. Ishak, C. Bogh, J. P. Grieco, N. L. Achee, and J. K. Baird. 2014. Impact of a spatial repellent on malaria incidence in two villages in Sumba, Indonesia. Am J Trop Med Hyg, 91: 1079-1087.

Swathi, S., G. Murugananthan, and S. K. Ghosh. 2010. Oviposition deterrent activity from the ethanolic extract of Pongamia pinnata, Coleus forshohlii, and Datura stramonium leaves against Aedes aegypti and Culex quinquefasciatus. Pharmacogn Mag, 24: 320-322.

Thanispong, K., N. L. Achee, M. J. Bangs, J. P. Grieco, W. Suwonkerd, A. Prabaripai, and T. Chareonviriyaphap. 2009. Irritancy and repellency behavioral responses of three strains of Aedes aegypti exposed to DDT and alpha-cypermethrin. J Med Entomol, 46: 1407-1414.

Tietze, N. S., K. R. Shaffer, and P. G. Hester. 1996. Half-life of naled under three test scenarios. J Am Mosq Control Assoc, 12: 251-254.

Trapido, H. 1954. Recent experiments on possible resistance to DDT by Anopheles albimanus in Panama. Bull World Health Organ, 11: 885-889.

Trexler J. D., C. S. Apperson, and C. Schal. 1998. Laboratory and field evaluations of oviposition responses of Aedes albopictus and Aedes triseriatus (Diptera: Culicidae) to oak leaf infusions. J Med Entomol, 35: 967-976.

Ujihara, K., T. Mori, and N. Matsuo. 2012. Recent advances of pyrethroids for household use. Top Curr Chem, 314: 31-48.

Unlu, I., M. A. Baker, N. Indelicato, D. Drews, Z. Zeng, and R. Vaidyanathan. 2018. Nighttime applications of two formulations of pyrethroids are effective against diurnal Aedes albopictus. J Am Mosq Control Assoc, 34: 158-162. DOI: 10.2987/17-6720.1

Unlu I., A. Farajollahi (Faraji), R. Ilia, T. N. Crepeau, S. Daniel, and R. Gaugler. 2014. Differences in male–female ratios of Aedes albopictus (Diptera: Culicidae) following ultra-low volume adulticide applications. Acta Trop, 137: 201–205. DOI: 10.1016/j.actatropica.2014.05.009

191

Unlu, I., A. Farajollahi (Faraji), D. Strickman, and D. M. Fonseca. 2013. Crouching tiger, hidden trouble: urban sources of Aedes albopictus (Diptera: Culicidae) refractory to source-reduction. PLoS ONE, 8: e77999. DOI: 10.1371/journal.pone.0077999

(USGS) U. S. Geological Survey. 2017. Rain and Precipitation, The USGS Water Science School. https://water.usgs.gov/edu/earthrain.html

Wagman, J. M., N. L. Achee, J. P. Grieco. 2015a. Insensitivity to the spatial repellent action of transfluthrin in Aedes aegypti: A heritable trait associated with decreased insecticide susceptibility. PLoS Negl Trop Dis, 9: 3726.

Wagman, J. M., J. P. Grieco, K. Bautista, J. Polanco, I. Briceo, R. King, and N. L. Achee. 2015b. The field evaluation of a push-pull system to control malaria vectors in Northern Belize, Central America. Malaria J. 14: 184.

Wilson, M. E. and L. H. Chen. 2015. Dengue: update in epidemiology. Curr Infect Dis Rep, 17: 457.

(WHO) World Health Organization. 2009. Dengue: guidelines for diagnosis, treatment, prevention and control. WHO Press, Geneva, Switzerland.

(WHO) World Health Organization. 2015. Global Malaria Programme, World Health Organization World Malaria Report 2015. WHO Press, Geneva, Switzerland.

Xue, R. D., A. Ali, and D. R. Barnard. 2004. Effects of forced egg-retention in Aedes albopictus on adult survival and reproduction following application of DEET as an oviposition deterrent. J Vec Ecol, 30: 45-48.

Xue, R. D., D. R. Barnard, and A. Ali. 2001. Laboratory and field evaluation of insect repellents as oviposition deterrents against the mosquito Aedes albopictus. Med Vet Entomol, 15: 126-131.

Xue, R. D., D. R. Barnard, and A. Ali. 2003. Laboratory evaluation of 18 repellent compounds as oviposition deterrents of Aedes albopictus and as larvicides of Aedes aegypti, Anopheles quadrimaculatus, and Culex quinquefasciatus. J Am Mosq Control Assoc, 19: 397-403.

Xue, R. D., D. R. Barnard, and A. Ali. 2006. Laboratory evaluation of 21 insect repellents as larvicides and as oviposition deterrents of Aedes albopictus (Diptera: Culicidae). J Am Mosq Control Assoc, 22: 126-130.

Xue, R. D., W. A. Qualls, J. Phillips, and T. Y. Zhao. 2012a. Insecticidal activity of five commercial mosquito coils against Anopheles albimanus, Aedes albopictus, and Culex quinquefasciatus. J Am Mosq Control Assoc, 28: 131-133.

192

Xue, R. D., W. A. Qualls, M. L. Smith, J. R. Weaver, M. K. Gaines, and M. Debboune. 2012b. Field evaluation of the OFF! Clip-on mosquito repellent (Metofluthrin) against Aedes albopictus and Aedes taeniorhynchus (Diptera: Culicidae) in northeastern Florida, J Med Entomol, 49: 652-655.

Yamada T., Hiroko K., B. G. Lake, and S. Kawamura. 2015. Lack of effect of metofluthrin and sodium phenobarbital on replicative DNA synthesis and Ki-67 mRNA expression in cultured human hepatocytes. Toxicol Research, 4: 901. DOI: 10.1039/c4tx00217b

Yamada, T., S. Uwagawa, Y. Okuno, S. M. Cohen, and H. Kaneko. 2009. Case study: an evaluation of the human relevance of the synthetic pyrethroid metofluthrin-induced liver tumors in rats based on mode of action. Toxicol Sci, 108: 59-68. DOI: 10.1093/toxsci/kfp007

Yu, K. X., C. L. Wong, R. Ahmad, and I. Jantan. 2015. Mosquitocidal and oviposition repellent activities of the extracts of seaweed Bryopsis pennata on Aedes aegypti and Aedes albopictus. Molecules, 20: 14082-14102. DOI: 10.3390/molecules200814082

Yuan, J. Z. and J. B. Huang. 2014. Research progress in application of metofluthrin for mosquito control. Chinese J Hyg Insectic Equip, 20: 82-87.

193

BIOGRAPHICAL SKETCH

Christopher Stephen Bibbs was born into an Air Force family. After growing up in the

Florida panhandle, Chris began pharmacy school and worked as an in-patient hospital pharmacy technician from 2006–2010. While working in medical care, his fascination with the biochemistry and cultural stigma of venomous arthropods turned him on to entomology.

Following his childhood dreams of playing with bugs, Chris changed majors to graduate from the honors curriculum of the Gulf Coast State College and Florida State University connect program and received his associate’s degree in the spring of 2010.

After transferring to the University of Florida Entomology and Nematology Department,

He committed his upper division baccalaureate studies to biosecurity and medical entomology.

Under the tutelage of Dr. Rebecca Baldwin, Chris supported UF by recruiting students, performing community outreach services, building digital resources for the entomology and nematology department, working in the UF/IFAS Extension Nematode Assay Laboratory, and performing research on a fungal pathogen of the invasive brown widow spider, Latrodectus geometricus. He received his bachelor’s degree in the spring of 2012.

Chris was then recruited by Dr. Dawn Gouge at the University of Arizona to join the public health entomology team as a graduate student. While at UA, he built partnerships with the

Tucson Poison and Drug Information Center; the Venom Immunochemistry, Pharmacology and

Emergency Response Institute; and Rare Disease Therapeutics, Inc. in order to document the urban ecology and management of the neurotoxic Arizona bark scorpion, Centruroides sculpturatus, in school systems and housing tracts. Chris also assisted the western extension network as a state insect diagnostician and curatorial assistant in the UA Insect Collection-

Southwestern Regional Collection Museum until he received his master’s degree in 2014.

194

Upon graduating, Chris moved to St. Augustine, FL to be an education specialist with the

Anastasia Mosquito Control District of St. Johns County (AMCD). After two years performing a combination of applied research, public health outreach, public schools teaching, and public relations, he laterally transitioned to the applied research department of AMCD and started an employee-degree program to study insect toxicology, behavior, and public health vector management. While simultaneously working as the biologist for the AMCD and being mentored in public health entomology by Dr. Phillip Kaufman, Chris received his doctorate in the summer of 2019. He ultimately joined Central Life Sciences (Central Garden and Pet) as a research and development entomologist for managing arthropod pests and vectors.

195