A SEARCH for SOURCES of NONPERSISTENT VIRUS VECTORS and RESERVOIRS at LOCAL and REGIONAL SCALES Gina Marie Angelella Purdue University

Purdue University Purdue e-Pubs

Open Access Dissertations Theses and Dissertations

January 2015 TRACKING PLANT VIRUS INFECTIONS THROUGH MULTIPLE DIMENSIONS: A SEARCH FOR SOURCES OF NONPERSISTENT VIRUS VECTORS AND RESERVOIRS AT LOCAL AND REGIONAL SCALES Gina Marie Angelella Purdue University

Follow this and additional works at: https://docs.lib.purdue.edu/open_access_dissertations

Recommended Citation Angelella, Gina Marie, "TRACKING PLANT VIRUS INFECTIONS THROUGH MULTIPLE DIMENSIONS: A SEARCH FOR SOURCES OF NONPERSISTENT VIRUS VECTORS AND RESERVOIRS AT LOCAL AND REGIONAL SCALES" (2015). Open Access Dissertations. 1084. https://docs.lib.purdue.edu/open_access_dissertations/1084

This document has been made available through Purdue e-Pubs, a service of the Purdue University Libraries. Please contact [email protected] for additional information. Graduate School Form 30 Updated 1/15/2015

PURDUE UNIVERSITY GRADUATE SCHOOL Thesis/Dissertation Acceptance

This is to certify that the thesis/dissertation prepared

By Gina M. Angelella

Entitled TRACKING PLANT VIRUS INFECTIONS THROUGH MULTIPLE DIMENSIONS: A SEARCH FOR SOURCES OF NONPERSISTENT VIRUS VECTORS AND RESERVOIRS AT LOCAL AND REGIONAL SCALES

For the degree of Doctor of Philosophy

Is approved by the final examining committee:

Ian Kaplan Chair Jeffrey Holland

Christian Krupke

Joseph Anderson

To the best of my knowledge and as understood by the student in the Thesis/Dissertation Agreement, Publication Delay, and Certification Disclaimer (Graduate School Form 32), this thesis/dissertation adheres to the provisions of Purdue University’s “Policy of Integrity in Research” and the use of copyright material.

Approved by Major Professor(s): Ian Kaplan

Steve Yaninek 12/9/2015 Approved by: Head of the Departmental Graduate Program Date TRACKING PLANT VIRUS INFECTIONS THROUGH MULTIPLE DIMENSIONS: A SEARCH FOR SOURCES OF NONPERSISTENT VIRUS VECTORS AND RESERVOIRS AT LOCAL AND REGIONAL SCALES

A Dissertation

Submitted to the Faculty

Purdue University

Gina M. Angelella

In Partial Fulfillment of the

Requirements for the Degree

Doctor of Philosophy

December 2015

Purdue University

West Lafayette, Indiana

For little Nadia

iii

ACKNOWLEDGEMENTS

I would like to thank J. Anderson, C. Krupke, and J. Holland for their guidance and helpful comments regarding experimental design and chapter edits. C. Williams, J. Namachek, and D. Egel also collaborated on Chapter 1, V. Nalam, P. Nachappa, and J. White on Chapter 4, and A. Michel on Chapter 3. I would also like to thank the numerous extension agents, private pumpkin and alfalfa growers, and the staff at Purdue University’s Meigs Farm who made specimen collections possible. C. Blubaugh, J. Carillo, M. Garvey, L. Ingwell, E. Long, P. Olaya, and E. Rowen gave helpful comments for chapter edits. Chapter 4 is dedicated to the memory of Peter Saya, whose EPG expertise was instrumental in initiating experimental assays. C. Williams, J. Nemachek, and D. Egel collaborated on Chapter 1, V. Nalam, P. Nachappa, and J. White collaborated on Chapter 4, and A. Michel on Chapter 3. J. White and A. Dehnel provided Aphis craccivora colonies and rearing expertise, D. Lagos and D. Voegtlin provided aphid identification expertise and training, S. Pearce, D. Egel, and S. Hoke assisted with sample collection, C. Michel with RNA extraction, S. Nouri and M. Deb provided advice regarding RNA extraction and/or primer design, B. Peterson assisted with nucleic acid extraction techniques, and J.L. White provided technical assistance with LINUX and STACKS installation. Funding support was provided by the USDA North Central Region IPM Grants Program, grant 11-34103-30723, the Indiana Vegetable Growers Association Grant Program, the Indiana Academy of Sciences Professional Grant Program, and NIFA USDA Predoctoral Fellowship grant 107483.

TABLE OF CONTENTS

Page

ABSTRACT ...... vi

CHAPTER 1 ...... 1

Introduction ...... 1

Methods ...... 3

Results ...... 7

Discussion ...... 17

References...... 22

CHAPTER 2 ...... 28

Introduction ...... 28

Methods ...... 29

Results ...... 32

Discussion ...... 35

References...... 39

CHAPTER 3 ...... 43

Introduction ...... 43

Page

Methods ...... 44

Results ...... 47

Discussion ...... 52

References...... 55

CHAPTER 4 ...... 58

Introduction ...... 58

Methods ...... 60

Results & Discussion ...... 62

References...... 67

APPENDICES

Appendix A ...... 70

Appendix B ...... 91

Appendix C ...... 96

VITA ...... 106

ABSTRACT

Angelella, Gina M. Ph.D., Purdue University, December 2015. Tracking Plant Virus Infections Through Multiple Dimensions: A Search for Sources of Nonpersistent Virus Vectors and Reservoirs at Local and Regional Scales. Major Professor: Ian Kaplan.

My dissertation explores the ecology of aphid-vectored viruses in Midwestern cucurbits; in particular, it focuses on identifying source populations of vectors and virus reservoirs within fields and interactions of vector aphids with land cover surrounding and within fields. Initially, I identified the most commonly occurring viruses and aphid species associated with virus infections in pumpkin fields located across Indiana. This was done by assaying cucurbit leaf tissue with multiplex-rt-PCR targeting all four aphid-vectored, nonpersistent viruses found in cucurbits (papaya ringspot virus type-W, watermelon mosaic virus type-2, zucchini yellow mosaic virus, cucumber mosaic virus) and concurrently monitoring aphid species alightment in fields throughout Indiana. Watermelon mosaic virus (WMV) was the most common infection, detected in all but one field across both years. Papaya ringspot virus (PRSV) was also detected in many fields, but not zucchini mosaic virus or cucumber mosaic virus. Therioaphis trifolii and Aphis craccivora were positively associated with PRSV, and Aphis nerii with WMV. Rhopalosiphum padi was negatively associated with WMV. No relationship between total aphid or noncolonizing species counts was found with virus infection, and an increase in total colonizing species (Aphis gossypii) trapped in fields predicted a decrease in PRSV infection. Negative relationships between aphid species alightment and virus infection in fields could result from a virus-induced reduction in pumpkin plant quality, thereby reducing abundance of a colonizing species within the field. Other virus-mediated effects in pumpkins could reduce aphid attraction to pumpkin fields as well, such as modification of the olfactory stimuli detectable by airborne aphids. I also inventoried weed species within fields and assayed samples for virus content, to investigate weed cover contributions to vector behavior or as host to virus or vector. Using land cover data from the National Land Cover Database, relationships among virus, aphids

vii

and land cover were also analyzed at spatial scales of 1–5 km distance surrounding focal pumpkin fields. A comparison among interactions within each spatial scale indicated relative predictive ability of vector-virus dynamics by surrounding land cover versus within-field weed cover, and the influential factors found at each scale. Although there was evidence for a negative interaction of vector species’ alightment with weed cover, surrounding landscape was consistently a stronger predictor of vector alightment within fields. My focus then narrowed to one aphid species, the cowpea aphid (Aphis craccivora Koch), implicated in virus spread. I collected the species from local and regional source populations (alfalfa and black locust seedlings) and used SNPs to assess their population genetics, and used these data to infer the dispersal patterns of cowpea aphids surrounding focal cucurbit fields, both identifying host-associated source populations and patterns in population structure by spatial distance. Most aphids grouped in one of two host-associated multilocus genotypes (MLGs), and temporal variation in focal field visitation by aphids indicated activity occurred earlier in the season in black locust-associated aphids than in alfalfa-associated aphids. Results also suggest local movement (<10 km) characterizes most cowpea aphid dispersal. Lastly, I explored host plant-associated endosymbiont effects on cowpea aphid feeding behavior in pumpkins. Interaction effects among host-association (locust/alfalfa), endosymbiont association (Arsenophonus/Hamiltonella defensa/cured) and virus infections (watermelon mosaic virus) on pumpkins were studied using an electrical penetration graph. I found differential endosymbiont effects on feeding behavior, with one exhibiting depressed frequency of intracellular probes and the other increased frequency. Greater probing frequency on WMV- infected pumpkins occurred across all aphids as a whole. This supports the Vector Manipulation Hypothesis, in which plant viruses mediate changes in plant gustatory and olfactory cues to change vector behavior in ways that enhance likelihood of virus transmission (i.e., increased exploratory intracellular probes). Overall, the dissertation results fill in some of the knowledge gaps existing in aphid-vectored, nonpersistent virus epidemiology using the pumpkin crop system. Hopefully some of the data will prove useful in applications toward aphid-vectored virus management in crops.

CHAPTER 1. DIFFERENTIAL LIFE HISTORY TRAIT ASSOCIATIONS OF APHIDS WITH NONPERSISTENT VIRUSES IN CUCURBITS

1.1 Introduction At least 275 plant viruses are transmitted by aphids (Nault 1997), and the vast majority (ca. 76%) of these are nonpersistent in nature (Katis et al. 2007). The nonpersistent mode of transmission, characterized by the short time in which the virus may be transmitted after acquisition, makes elucidating virus epidemiology a challenge due to the rapid onset and transient nature of vector competence. Further, these viruses are often transmissible by many aphids, e.g., cucumber mosaic virus (CMV) can be vectored by >60 species (Douine et al. 1979). Due to the large number of potential vectors, viral epidemics can be differentially triggered by the action of colonizing species (i.e., those that feed and reproduce on the plant) or noncolonizing species (i.e., those that briefly land on plants and fly away, after conducting ‘taste’ probes preceding rejection). This dichotomy is critical because managing colonizing vectors is far easier than noncolonizers in agricultural environments; namely, colonizers can be controlled with insecticides, whereas noncolonizers cannot and, instead, management tactics are aimed at disrupting alightment (e.g., row covers, mulches, manipulating local vegetation; Broadbent 1957, Perring et al. 1989). Noncolonizers typically conduct more frequent shallow probes of peripheral tissues rather than feeding on phloem, thus increasing chances of pathogen acquisition or inoculation, and also move more frequently among nonhost plants than colonizers, potentially enhancing virus spread (Peters et al. 1990, Kanavaki et al. 2006). Indeed, several studies have implicated noncolonizing aphids as the most important vectors of nonpersistent viruses in crop systems (Raccah et al. 1985, Summers et al. 1990, Fereres et al. 1992, 1993, Webb et al. 1994, Perez et al. 1995, Nebreda et al. 2004). In certain cases, this is a result of noncolonizers exhibiting higher transmission efficiencies (Peters et al. 1990), but in other cases this is simply due to greater vector pressure (i.e., high numbers of aphids landing in fields). For example, the soybean aphid, Aphis glycines Matsumura, is a pest that recently invaded the Midwestern U.S., and has rapidly become the numerically dominant species in aphid community surveys (Nault et al. 2004, Mueller et al.

2010). Interestingly, the invasion and spread of soybean aphid in the Midwest coincided with increases in the incidence of aphid-vectored viruses in fruit and vegetable crops in this region since 2000, thereby implicating the soybean aphid as a vector in non-soybean systems (Davis et al. 2005, Davis & Radcliffe 2008, Gildow et al. 2008, Nault et al. 2009). This pattern is evident despite the fact that the soybean aphid is a soybean specialist. Although noncolonizers are ecologically distinct from colonizers, their influence on virus presence may not be mutually exclusive or the two groups may act in concert. Some analyses indicate that total aphid alightment, independent of colonizer/noncolonizer designation, is the best predictor of virus prevalence (Madden et al. 1987, Mora-Aguilera et al. 1992, DiFonzo et al. 1997, Dusi et al. 2000, Katis et al. 2006). In a related study of cowpea aphid-borne mosaic virus, noncolonizing aphids were important instigators of initial infections within fields, but pathogen prevalence among plants was correlated with the abundance of colonizers on plants (Atri et al. 1986). This suggests differential roles of noncolonizing vs. colonizing aphids in transporting virus into fields from outside reservoirs and spreading virus within fields, respectively. I investigated the role of colonizing and noncolonizing aphids on virus prevalence in pumpkins (Cucurbita pepo L., C. mixta Pang., C. maxima Dutch., or C. moschata Poir.), an important specialty crop in Indiana and the Midwestern U.S. in general. The four aphid-vectored pumpkin viruses in this region are: cucumber mosaic virus (CMV; Bromoviridae), papaya ringspot virus type W (PRSV; Potyviridae), watermelon mosaic virus type 2 (WMV; Potyviridae), and zucchini yellow mosaic virus (ZYMV; Potyviridae) (Zitter et al. 1996). In the Midwest, this complex of viruses was rated as both common and difficult to control by a consortium of specialists and growers (Paulsrud et al. 2005), and ultimately leads to reductions in farm revenue (Walters 2003). In a four year study of pumpkins in Southern Indiana, nearly 100% of plants were infected with virus by late-season, resulting in dramatic (>50%) reductions in yield (Brust 2000). This outcome matches reports from growers, one whom felt that he could not grow pumpkins in his county and thus rented fields elsewhere due to perennial virus pressure (personal communication; Liz Maynard, Extension Specialist, Purdue University). Because transmission occurs rapidly, insecticides are not an effective management tool, and alternative methods such as reflective plastic mulch are not cost effective or practical for large-scale or direct seeded operations (Paulsrud 2005). Current management recommendations include planting cucurbit crops far apart from each other to minimize virus spread (Brust & Everts 2010), and planting early and controlling weeds within and around fields (Egel 2013).

Although nonpersistent virus infections in pumpkins seem to be localized within the Midwest, it is unknown why, or which aphids are responsible for initial infections and secondary spread within fields. At least 80 aphid species have been reported flying in and around vegetable fields in this region (Kagezi et al. 1999), thus many species may be contributing as noncolonizing vectors. I conducted a two year survey on commercial pumpkin farms to assess the aphid community landing in fields and virus assemblage infecting pumpkins to evaluate the relationship between colonizing/noncolonizing aphid species and virus dynamics. I hypothesized that noncolonizers exert the greater influence on virus dynamics, and that the most plentiful noncolonizing aphid species would be the best predictors of virus occurrence. 1.2 Methods Over two years, aphid alightment and virus infection were monitored in commercial pumpkin fields ranging in size from 0.4–8 ha throughout Indiana (2010: n=13, 2011: n=16) (Fig. A1-1). In each field, I sampled leaf tissue to determine virus occurrence and deployed pan traps to inventory alighting aphids concurrently. Viruses of interest included the four aphid-vectored nonpersistent viruses infecting pumpkins: CMV, PRSV, WMV, and ZYMV. Aphid Alightment Five pan traps were spaced evenly throughout each field in accessible locations (close to roadways or the field edge) between pumpkin plants in a row. Pan traps consisted of 1.9 l clear, round Rubbermaid® bowls (Instawares Restaurant Supply, Kennesaw, GA), containing 0.5 l of a 1:4 solution of propylene glycol (Qualichem, Salem, VA) and water to preserve aphid catches, with a green ceramic tile (CoolTiles, Hicksville, NY) in the bottom of the pan to attract alates (DiFonzo et al. 1997). Pans were secured at canopy height on the top rung of a 107 cm modified and inverted tomato cage (Lowe’s, Lafayette, IN), and the cages securely anchored in the ground. Pan trap catches were collected once a week and alates preserved in 70% ethanol for identification. Trap monitoring was initiated in fields after seedling emergence or transplant on or after the week of June 30th, and ended the week of August 25th. In 2010, three fields were visited the initial week of June 30th, twelve the next week, and thirteen for the remaining weeks. In 2011, eleven fields were visited the initial week, fifteen the next week, and sixteen throughout the remaining weeks. A majority of alates were identified to species with a dissecting microscope, and the remainder slide-mounted. Identifications in 2010 were made under the supervision of Dr. David Voegtlin and Dr. Doris Lagos (Illinois Natural History Survey and Department of Entomology, University of Illinois at Urbana-Champaign). Dichotomous keys were also used to confirm aphid identification (Smith et al. 1992, Boydston and Allison 2003). Additionally, I

identified aphid species colonizing pumpkin plants using a hand lens, a subset of which was verified using a dissection scope. Weekly apterous/alate on-leaf aphid counts were taken in 2010 on 10 randomly-selected plants. Virus Assays Leaf tissue was sampled by walking through each field in a W-formation and randomly selecting plants, collecting one young terminal leaf per plant. Although growers are believed to have planted a large variety of cultivars, and the resistance/tolerance traits of these cultivars are unknown, this potential pitfall was hopefully overcome by the sample size and random sampling protocol within fields. Twenty leaves were collected per field at the end of the monitoring period, between August 19th and September 2nd in 2010, and between August 22nd and August 30th in 2011. Leaves were sealed in a plastic bag, transported on dry ice, and stored in a -80°C freezer until processed for viral content. Pooled samples from 2010 were evaluated as well (n=20 samples/field) from initial visits to each field starting June 30th (corresponding to the initiation of trap monitoring), to assess initial virus infection among fields. Prior studies of a similar nature have utilized larger leaf sample sizes (e.g., Katis et al. 2006, Nault et al. 2009, Mueller et al. 2012); however, the sample size of this study is comparable in sample number per hectare of field and/or in processing costs. Similar previous studies used enzyme linked immunosorbent assays (ELISA) to process samples, which has lower processing cost per sample than PCR, and had either much larger field sizes or a much smaller number of field sites. Leaf tissue was homogenized in liquid nitrogen prior to total RNA extraction with Purelink® RNA Mini Kit (Life Technologies, Carlsbad, CA) and treatment with DNaseI (New England BioLabs, Ipswich, MA). Approximately 250 mg of leaf tissue was used as starting material, either from one leaf with leaves assayed separately, or ca. 12.5 mg from each of the 20 leaves/field in pooled assays. First-strand cDNA was synthesized using a blend of oligo(dT) and random hexamer primers plus 2 µg of total RNA in a 20 µl final reaction volume, using iScript cDNA Synthesis Kit (Bio-Rad, Hercules, CA) according to the manufacturer’s protocol. A multiplex reverse transcriptase PCR (m-RT-PCR) procedure was developed using primers for CMV, WMV, ZYMV, and PRSV. Primers were designed by aligning nucleotide sequences from multiple isolates of each virus documented in NCBI using CLUSTALW software (Larkin et al. 2007), and analyzing conserved regions in primer BLAST to identify new potential primers. Primers targeting each virus were deemed useful upon successfully amplifying viral RNA from control samples, which were procured from ATCC (Manassas, VA): CMV (ATCC® PV-548™), PRSV (ATCC® PV-23™), WMV (ATCC® PV-27™), and ZYMV (ATCC® PV-595™). Primer

pair candidates with similar melting temperatures (TM) that amplified sufficiently distinct product sizes were combined to form m-RT-PCR primer pools, from which an effective set of multiplex primers was selected (Table A1-1). The m-RT-PCR procedure was carried out to determine presence/absence of virus using 3 µl of cDNA (>250 ng) and 0.5 µl of each 10 µM forward and reverse primer, added to 12.5 µl of GoTaq® Green Master Mix (Promega, Madison, WI) for a final volume of 25 µl. Samples were amplified using the C1000 Touch™ Thermal Cycler (Bio- Rad, Hercules, CA) under the following conditions: initial Taq polymerase activation at 95°C (15 min), followed by 29 cycles of denaturation at 94°C (1 min), annealing at 55°C (1 min), and extension at 72°C (1 min), and followed by a final extension at 72°C (10 min). I separated amplified fragments in a 2% agarose gel (Benchmark Scientific, South Plainfield, NJ) in 1xTris- Borate-EDTA buffer (Promega, Madison, WI) using gel electrophoresis (10V-1cm gel) and ethidium bromide stain, and viewed them under UV light with a gel documentation system (GelDoc Bioimaging System, UVP Inc., Upland, CA). Product bands were sized relative to the HyperLadder™ IV molecular weight marker (Bioline, Taunton, MA). Each analysis contained a no-template negative control, as well as a positive control containing virus RNA (ATCC, Manassass, VA). Statistical Analyses Data were analyzed separately by year. I performed arcsine-square root transformations to normalize the proportion of PRSV- and WMV-infected samples prior to analysis. (CMV and ZYMV were never detected and thus not analyzed.) Aphid pan trap counts were converted to aphid-days for analyses by calculating the area under the insect trap-count curve:

, where x =avg. species count trap-1 field-1in sample week i, and i di=first day in sampling week i (Hanafi et al. 1989). This generates total cumulative alightment values for each field that have been corrected for any variation in trap number or duration occurring throughout a season. The statistical software program R version 2.15.2 was used to conduct analyses (2011). Virus presence/absence among fields. If variation was found in the presence or absence of a virus species among fields, aphid species data were converted to Bray-Curtis dissimilarity matrices (Legendre & Legendre 1998), and permutational multivariate analyses of variance (PERMANOVAs) were performed to assess whether aphid community composition varied with virus presence or absence (Anderson 2001). Prior to PERMANOVA, the 2010 aphid community data were standardized, due to exceptionally large melon aphid abundance relative to all other species (Table A1-2), and converted to a Euclidean dissimilarity matrix, and the 2011 aphid

community data were converted to a Bray-Curtis dissimilarity matrix (Legendre & Legendre 1998). Bray-Curtis is an excellent dissimilarity metric for use in community structure analyses, but data standardization created negative values which would render the Bray-Curtis metric meaningless, which is why Euclidean was used instead on 2010 aphid community data (Legendre & Legendre 1998). Next, the assumption of homogeneity of group dispersion was tested on unstandardized aphid community data by conducting an ANOVA on the distances between group (virus presence or absence) members and group centroids (Anderson 2006). Similarity percentages (SIMPER) analyses were then performed on aphid abundance data to identify the aphid species contributing the largest percentage variation in aphid community structures by virus presence or absence by year, assessed using Bray-Curtis dissimilarity matrices (Clarke 1993). The aphid species identified by SIMPER analysis as contributors of >95% variation in aphid community structure between fields grouped by virus presence vs. absence were subjected to t tests, to examine whether mean species alightment data varied between infected and uninfected fields, and logistic regressions performed to examine the degree to which each aphid species predicts the probability of virus presence or absence among fields (Amemiya 1985). To examine the possible role of aphid host-plant preference in relation to presence/absence of a virus species in fields, total noncolonizers (those aphids which do not feed and reproduce on pumpkins) and melon aphids (Aphis gossypii Glover, pumpkin colonizers) were assessed separately using logistic regressions. Logistic regressions also tested the predictive power of total aphid alightment for estimating virus presence/absence probability among fields. T tests assessed differences in the mean alightment of total aphids by virus presence or absence. Virus prevalence within fields. Assessments of the best-fitting model of aphid vectors related to virus prevalence, or the proportion of virus-infected samples within fields, were generated with an exhaustive, all-subsets regression analysis. Aphid community data from 2010 were again standardized prior to analyses. This selected the best-fitting model of each size (number of predictors per equation) containing aphid species predictor variables to describe the proportion of infected samples per field (Miller 2002). Of these, the model with the lowest corrected Akaike Information Criterion value (AICc) was selected as the best predictor of the proportion of virus-infected samples among fields (Burnham and Anderson 2002). The extent to which each aphid species in the model predicted virus prevalence in fields was assessed individually using bivariate linear regressions (Kennedy and Keeping 1962). Total noncolonizer and colonizer relationships with virus prevalence were assessed separately using bivariate linear

regressions. I conducted linear regressions of total, cumulative aphid alightment and proportion of infected samples to assess how well total aphid alightment predicted virus prevalence. Early season aphid alightment. Although the chronology of virus infection events remains unknown, assessments were also conducted relating virus prevalence or presence/absence in fields with aphid community data from early in the sampling season (weeks 1 + 2). This was done for several reasons: first, early growing season is a critical time for influencing pumpkin crop yield, as the greatest reductions occur with early infections prior to fruit set (e.g., Fletcher et al. 2000); second, it is well known that the degree of contrast between foliage and bare ground – markedly greater for young plants early in the growing season – mediates aphid alightment behavior, with many species exhibiting greater preference for high contrast (e.g., Favret & Voegtlin 2001). Indeed, cultural management practices reducing contrast within fields have been shown to reduce the risk of aphid-vectored nonpersistent virus infection in pumpkins (Brust 2000). As such, to assess the relationship of virus prevalence and early total aphid, noncolonizer, and colonizer alightment, as well as alightment of each aphid species, linear regressions were conducted. Likewise, virus presence or absence was related to each of these independent variables with logistic regressions and t-tests. Aphid species present in ≤2 fields were excluded from analyses. 1.3 Results At least 53 aphid species, from 29 genera, were identified in 2010 and 2011, with melon aphid representing the largest trap catch in both years (Fig. 1-1, Table A1-2). The total number of noncolonizing aphids collected was fairly consistent between years, with Rhopalosiphum padi (L.) (bird cherry-oat aphid) the numerically dominant noncolonizer in 2010, and Aphis craccivora Koch (cowpea aphid) followed by Therioaphis trifolii (Monell) f. maculata (spotted alfalfa aphid) preponderant in 2011 (Fig. 1-1, Table A1-2). On-leaf aphid counts were not used in analyses, as they were highly correlated with melon aphid pan trap counts (R2 = 0.81; F = 29.44; df = 1,7; RSE = 0.17; P < 0.001; Fig. A1-2). Thus, melon aphid pan trap data used in analyses also reflect abundance on plants.

(A) Noncolonizers: (B) Colonizers: 2010 2010 120 16

1 100

14 -

1 - 12 80 10 8 60 6 40

4 weekly count field count weekly

weekly count field count weekly 20 2 0 0

2011 9 2011 14

1 -

1 12

7 - 6 10 5 8 4 6 3

2 4 weekly counts field counts weekly 1 field counts weekly 2 0 0

other noncolonizers cowpea aphid spotted alfalfa aphid bird cherry-oat aphid melon aphid Figure 1-1. Weekly pan trap counts per field of preponderant noncolonizing and colonizing species in 2010 and 2011: mean ± se. A) Noncolonizers: the cowpea aphid, spotted alfalfa aphid, and other noncolonizing species. B) Colonizers: the melon aphid.

Early-season virus assays in 2010 detected WMV in only a small fraction of fields (2/13), and PRSV was not detected in any fields (0/13). At the end of the monitoring period in 2010, however, all fields were infected with WMV and many with PRSV (Table 1-1). Similarly, end of season assays in 2011 detected WMV in almost all fields and PRSV in the majority of fields (Table 1-1). Mixed infections were common in both years (Table A1-3). Neither CMV nor ZYMV were detected in any fields in either year.

Table 1-1. The fraction of virus-infected fields surveyed, and the mean proportion of inoculated pumpkin leaf samples (±SE) in 2010 & 2011. Fields with Virus Incidence/Total Mean Proportion of Samples Infected (SE)

Virus 2010 2011 2010 2011

WMV 13/13 15/16 0.69 (0.09) 0.55 (0.08)

PRSV 11/13 10/16 0.28 (0.09) 0.25 (0.06)

ZYMV 0/13 0/16 — —

CMV 0/13 0/16 — —

Pumpkin leaf samples (n=20) were collected in early Sept. from field sites (2010: n=13, 2011: n=16)

Virus-Aphid Interactions Because fields were not uniformly infected with PRSV, PERMANOVA and SIMPER analyses were conducted to examine variation in aphid community structure relative to PRSV presence or absence in fields. The dispersion of aphid communities grouped by PRSV presence or absence in a field were not significantly different in 2010 (F = 1.68; df = 1,11; P > 0.1) or in 2011 (F = 0.0006; df = 1,14; P > 0.1); thus, data from both years met the assumption of homogeneity of variances necessary for conducting a PERMANOVA (Anderson 2006). Variation in aphid community structure was significantly associated with PRSV presence/absence in 2011 (R2 = 0.15; F = 2.49; df = 1,14; P < 0.05), but not 2010 (R2 = 0.045; F = 0.52; df = 1,11; P > 0.1). Due to nearly uniform infection status of WMV across all field sites in 2010 and 2011 (Table 1-1), only WMV prevalence (the proportion of infected WMV samples within fields) was utilized as the response variable. Colonizers vs. noncolonizers. Neither noncolonizer nor colonizer alightment totals were significantly and positively associated with proportions of virus-infected samples within fields. Only colonizer (melon aphid) alightment explained variation in PRSV-infected samples among fields in 2011 (y=0.59+0.0053x; RSE = 0.32; R2 = 0.32; F = 6.70; P < 0.05; Table 1-2); however, it showed a negative relationship with PRSV (Fig. 1-2). Likewise, linear regressions of colonizing and noncolonizing aphid species yielded one significant negative relationship between noncolonizers and WMV in 2010 (y=0.54+0.00048x; RSE = 0.42; R2 = 0.35; F = 1.75; P < 0.05) (Fig. 1-3; Table 1-2), which was likely driven by the numerically-dominant bird cherry-oat aphid. SIMPER analyses of PRSV presence/absence and aphid alightment showed that in both years the melon aphid contributed the greatest to overall community dissimilarity between PRSV-infected and uninfected fields: 42% in 2010, and 29% in 2011 (Table 1-3). In spite of this, melon aphid abundance was not a significant predictor of PRSV virus presence/absence in fields either year (Table A1-4a,b).

Table 1-2. Relationship of PRSV or WMV with colonizer (melon aphid) vs. noncolonizer (all other) species and variation in proportion of infected samples among fields, analyzed with linear regression. A) 2010. B) 2011.

A) 2010:

Virus Species Regression formula Residual SE R2 F P

PRSV Colonizer y=0.43+0.00029x 0.39 0.16 2.20 >0.1

Noncolonizers y=0.54+0.00048x 0.43 0.0042 0.05 >0.1

Colonizer y=0.99+0.00028x 0.44 0.13 1.62 >0.1 WMV Noncolonizers y=0.54+0.00048x 0.35 0.44 1.75 <0.05

B) 2011:

Virus Species Regression formula Residual SE R2 F P

Colonizer y=0.59+0.0053x 0.32 0.32 6.70 <0.05 PRSV Noncolonizers y=0.31+0.0020x 0.36 0.11 1.75 >0.1

Colonizer y=0.95-0.0035x 0.42 0.11 1.68 >0.1 WMV Noncolonizers y=0.80+0.00088x 0.44 0.015 0.216 >0.1

Table 1-3. SIMPER analysis of aphid species per fields grouped by PRSV presence (P) or absence A). Aphid species listed in descending order by contribution to community composition differences between uninfected and virus-infected fields. A) 2010: Species Avg. abun. P Avg. abun. A Contrib. % Cum. %

Aphis gossypii 287.8 146.7 42.0 61.2

Rhopalosiphum padi 42.6 11.6 14.7 82.6

Aphis glycines 1.3 4.9 2.5 86.2

Therioaphis trifolii 3.9 4.9 1.9 89.1

Rhopalosiphum maidis 3.9 4.6 1.6 91.4

Aphis craccivora 5.4 1.1 1.5 93.5

Hyadaphis foeniculi 1.7 0.4 0.7 94.5

Tetraneura spp. 0.4 1.1 0.6 95.3

B) 2011: Species Avg. abun. P Avg. abun. A Contrib. % Cum. %

Aphis gossypii 12.0 60.9 29.2 45.9

Aphis craccivora 32.3 10.9 9.4 60.6

Therioaphis trifolii 10.6 4.1 4.9 68.4

Tetraneura spp. 2.6 4.2 3.3 73.6

Pemphigus spp. 3.0 3.6 2.3 77.2

Rhopalosiphum padi 2.3 2.8 2.1 80.5

Rhopalosiphum maidis 1.4 2.2 1.5 82.8

Aphis nerii 1.7 1.4 1.3 84.9

Capitophorus elaeagni 0.7 1.6 1.2 86.8

Anoecia spp. 1.2 0.5 1.0 88.4

Colopha ulmicola 0.9 0.7 1.0 90.0

Aphis fabae 0.6 1.1 1.0 91.6

Aphis glycines 0.6 1.2 0.8 92.9

Uroleucon spp. 0.0 0.8 0.8 94.1

Acyrthosiphon pisum 1.1 0.4 0.7 95.2

Avg. abund.: average abundance; contr. %: average contribution percentage to overall dissimilarity; Cum %: cumulative percentage of contribution dissimilarity

Figure 1-2. Relationship between proportion (prop) of PRSV-inoculated samples and melon aphids or noncolonizing species of aphids (aphid-days, the average alightment count trap-1 in a field each week, summed across the sampling season), analyzed by linear regression on arcsine-square root transformed proportion data. A) 2010 prop PRSV and melon aphid. B) 2010 prop PRSV and noncolonizers. (C) 2011 prop PRSV and melon aphid. (D) 2011 prop PRSV and noncolonizers.

Figure 1-3. Relationship between proportion (prop) of WMV-inoculated samples and melon aphids or noncolonizing species of aphids are presented as in Figure 1-2. A) 2010 prop WMV and melon aphid. B) 2010 prop WMV and noncolonizers. (C) 2011 prop WMV and melon aphid. (D) 2011 prop WMV and noncolonizers.

Total aphids. There was no relationship between total aphid alightment and the proportion of PRSV-infected samples in 2010 (y=0.41+0.00027x; RSE = 0.40; R2 = 0.15; F = 1.89; P > 0.1) and 2011 (y=0.44-0.00018x; RSE = 0.39; R2 = 0.0012; F = 0.017; P > 0.1) (Fig. A1-3a,b), or with WMV-infected samples in 2010 (y=0.99+0.00021x; RSE = 0.45; R2 = 0.079; F = 0.95; P > 0.1) and 2011 (y=0.88-0.00044x; RSE = 0.44; R2 = 0.0058; F = 0.082; P > 0.1; Fig. A1-3c,d). Thus, total aphid alightment did not predict either PRSV or WMV prevalence within

fields. Further, logistic regression of PRSV presence/absence by total aphids was not significant in either year (2010: y=1.52+0.00067x; RSD = 11.01; z = 0.33; P >0.1; 2011: y=0.70-0.0022x; RSD = 21.07; z = -0.32; P > 0.1; Fig. A1-4), nor were t tests of total aphid mean alightment by PRSV presence/absence in fields (2010: t = -0.67; df = 8.00; P > 0.1; 2011: t = 0.33; df = 13.36; P > 0.1; Fig. A1-4), showing that total aphid alightment neither differed between PRSV-infected and uninfected fields, nor could it predict PRSV infection among fields. Individual aphid species. The best-fitting model of aphid vectors related to virus prevalence, or the proportion of PRSV-infected samples within fields, included abundance of Tetraneura spp., soybean aphid (A. glycines Matsumura), and greenbug [Schizaphis graminum (Rondani)] in 2010 (Table A1-5a), and the corn leaf aphid [Rhopalosiphum maidis (Fitch)], honeysuckle aphid [Hyadaphis foeniculi (Passerini)], melon aphid, soybean aphid (A. glycines Matsumura), cowpea aphid, and bean aphid (Aphis fabae Scopoli) in 2011 (Table A1-5b). None of these aphids individually predicted PRSV prevalence in fields (Table A1-6). Although the melon aphid was the greatest contributor to overall dissimilarity among aphid communities grouped by PRSV presence or absence in fields, 7 additional species drove community dissimilarity in 2010, and 14 additional species in 2011 (Table A1-4). The bird cherry-oat aphid was the second highest contributor in 2010 at 15%. In 2011, the cowpea aphid was the second highest contributor at 9%, followed by the spotted alfalfa aphid at 5%. Of all SIMPER-identified contributors, only cowpea aphid mean alightment varied in fields by infection status in 2010 (t = -2.59; df = 5.27; P < 0.05; Fig. 1-4B), and spotted alfalfa aphid mean alightment in 2011 (t = - 3.12; df = 13.67; P < 0.01; Fig. 1-4D). No aphid species exhibited a relationship (α < 0.05) with PRSV infection status in 2010 or 2011 after logistic regression, but in 2011 both melon aphid (y=1.92-0.051x; RSD = 14.23; z = -1.74; P = 0.08) and spotted alfalfa aphid (y=0.31x-1.68; RSD = 14.90; z = 1.92; P =0.06) (Fig. 1-4C) were marginally significant (Table A1-4).

Figure 1-4. Relationship between PRSV presence in fields and spotted alfalfa aphid or cowpea aphid total cumulative alightment in fields (aphid-days, the average alightment count trap-1 in a field each week, summed across the sampling season). Logistic regressions (panels A and C): =PRSV presence (1) or absence (0) in fields by aphid alightment, = fitted regression values of estimated PRSV odds probability given aphid alightment in a field. Boxplots (panels B and D): aphid alightment in fields grouped by PRSV presence or absence in fields, displaying median values and interquartile ranges, and pairwise group comparisons tested with student’s t test. A) 2010 logistic regression of PRSV presence probability and cowpea aphid. B) 2010 median cowpea aphid alightment by PRSV presence or absence. (C) 2011 logistic regression of PRSV presence probability and spotted alfalfa aphid. (D) 2011 median spotted alfalfa aphid alightment by PRSV presence or absence.

The best fit model for 2010 included abundance of bird cherry-oat aphid, soybean aphid, pea aphid, and bean aphid (Table A1-7a). In 2011, the model contained bird cherry-oat aphid, potato aphid [Macrosiphum euphorbiae (Thomas)], honeysuckle aphid, Anoecia spp., oleander

aphid, Aphis lacinariae Gillette & Palmer, and pea aphid (Table A1-7b). Bivariate linear regressions yielded only one significant interaction in each year. In 2010, presence of bird cherry- oat aphid explained variation in proportion of WMV-infected pumpkins (y=1.35-0.0070x; RSE = 0.34; R2 = 0.47; F = 9.70; P < 0.01; Fig. 1-5A; Table A1-8), but exhibited a negative relationship with WMV infection. In 2011, the oleander aphid exhibited a positive relationship with WMV (y=0.67+0.11x; RSE = 0.38; R2 = 0.28; F = 5.41; P < 0.05; Fig. 1-5B; Table A1-8).

Figure 1-5. Relationship between proportion (prop) of WMV-inoculated samples and the bird cherry-oat aphid or the oleander aphid are presented as in Figure 2. A) 2010 prop WMV and bird cherry-oat aphid. B) 2010 prop WMV and the oleander aphid.

Early season. Because only one melon aphid was recorded during weeks 1 and 2 of 2011, early colonizer alightment was not analyzed in relation to 2011 virus data but can be assumed not a reliable predictor of virus dynamics within 2011 sample sites. Neither total aphid, colonizer, nor noncolonizer alightment rates were significant predictors of either PRSV or WMV prevalence (Table A1-9), or of PRSV occurrence within fields (Table A1-10) in 2010 or 2011 (Table A1-9). Only one species exhibited a significant relationship with virus prevalence: the bird cherry-oat aphid was negatively associated with WMV prevalence in 2010 (Table A1-9). In 2010, mean alightment rates of both colonizers and cowpea aphids trended toward significant difference by PRSV infection status (Table A1-10). Early season colonizers were recorded in 4/11 and cowpea aphids in 5/11 fields subsequently testing positive for PRSV at the end of the season, while neither species was recorded in uninfected fields. Although both colonizers and cowpea aphids exhibited low alightment rates during weeks 1 and 2, means were marginally greater in PRSV-

infected than uninfected fields (Fig. A1-5). No other significant differences were found among species’ alightment rates in either year (Table A1-10). 1.4 Discussion Virus Incidence Based on similar studies in the Midwestern U.S., it was not surprising that WMV was most frequently detected. Earlier surveys of cucurbit viruses in Illinois and Indiana also found WMV to be the most prevalent, detected from ca. 50% to >80% of samples (Jossey and Babadoost 2008). Unlike these studies, however, I detected PRSV at a much higher frequency and found no evidence for CMV or ZYMV. This could be due in part to my sampling regime, which only included end of the season leaf tissue collections after PRSV infections had more time to spread. Additionally, prior studies used ELISA to detect viruses, whereas PCR is >100-times more sensitive (Hu et al. 1995). It is also possible that CMV and ZYMV infections were present, but that the primers developed were ineffective at targeting endemic isolates. This seems unlikely given that the primers were designed to target highly conserved genomic regions among subtypes of each virus species. Indeed, samples collected in 2012 from a Southern Indiana pumpkin field and assayed with the multiplex-RT-PCR procedure described here tested positive for CMV, suggesting that the primer set effectively detects endemic isolates (Angelella, personal observation). The lack of virus in early-season assays suggests that seed transmission was not a factor underlying PRSV or WMV incidence, corroborating previous studies that also found no evidence for seed-mediated transmission with these two viruses (Wakman et al. 2002, Coutts et al. 2012). Rather, viruses were detected with subsequent assays conducted at the end of the growing season, after the occurrence of aphid flights. This suggests viruses originated from extra-field sources (e.g., weed reservoirs, neighboring crops) and implicates aphids as the causal vectors (although mechanical inoculation via farm equipment is another possible route, e.g., Coutts et al. 2012, 2013). Total Aphids Despite the evidence for aphids as virus vectors in this system, I found no relationship between total aphid alightment and virus presence/absence or within-field prevalence for either of the two viruses in each of two years. Thus, my data strongly indicate that aphids, as a group, offer no predictive power in forecasting virus-crop dynamics. This was surprising because nonpersistent viruses can be transmitted by a wide range of aphids, making total aphid catch a seemingly intuitive proxy for predicting viral epidemics. For example, 16 of 19 species tested

were capable of transmitting ZYMV in zucchini, albeit at varying efficiencies (Katis et al. 2006). It may be that total vector count is too simplistic as a single explanatory variable for plant virus pathosystems that are notoriously complex. Comparing aphid communities within virus-infected and uninfected fields provided further support for this notion: whereas total aphid counts could not predict field infection status, the composition of aphid species found within field communities could. A major assumption in summing aphids across the entire community is that vector competence, as determined by simple transmission efficiency tests, is a key trait explaining vector success. In reality, this only plays one part in an organism’s vectorial capacity, which must also take into account fundamental differences in vector ecology. The potential to transmit means very little if those species never come in contact with the virus. A more critical factor may be aphid relationships in the greater agroecosystem with plants that serve as virus reservoirs. Unfortunately, this is far less tractable due to variation in aphid diet breadth and host-plant use. Recent CMV outbreaks in snap beans (Phaseolus vulgaris L.) in the Great Lakes region were associated with the spotted alfalfa aphid and pea aphid (Nault et al. 2004, 2009). The host plant source population of these species is likely alfalfa, a known CMV reservoir. In this case, the importance of these vectors compared with other aphids is putatively due to their unique association with a common reservoir plant, rather than sheer abundance or superior transmission efficiency. Moreover, perennial plants such as alfalfa act as a continual inoculum source for successive generations, rendering them especially effective reservoirs (e.g., Seabloom et al. 2009, Borer et al. 2010). This outcome matches anecdotal reports from pumpkin growers in Indiana, some of whom report site-specific virus incidence (i.e., fields in certain locations always seem prone to getting virus, whereas other fields rarely become infected), suggesting that the composition of local vegetation communities is central to predicting virus damage in crop environments. Colonizers vs. Noncolonizers Unlike total aphids, I found relationships involving components of the full community differentiated according to crop colonization status; namely, colonizing aphids were associated with PRSV in 2011 and noncolonizing aphids with WMV in 2010. In both cases, however, I detected negative relationships between aphid alightment and virus incidence. Because of the correlative nature of most virus/vector field studies, it becomes impossible to separate cause from consequence. I assume that changes in vector abundance cause differential virus infection, i.e., high aphid trap catch increases virus. With colonizing species, the reverse may be true, i.e., crop virus infection affects aphid abundance. I suspect this may explain why PRSV displayed such a

strong negative relationship with melon aphid. Nonpersistent virus infection is predicted to reduce plant quality for insect vectors (Mauck et al. 2012). For example, melon aphid field abundance and population growth were dramatically lower on CMV-infected squash compared with virus-free plants (Mauck et al. 2010). In my case, I predict that PRSV infection reduced pumpkin quality for melon aphids, which then reduced their reproductive rate and the number of individuals that subsequently landed in pan traps. However, as the impact of PRSV infection on pumpkin plant quality throughout the study is unknown, more substance to this speculation cannot be provided until melon aphid performance on PRSV-infected pumpkin is investigated. An alternative hypothesis is that PRSV infection reduced crop attractiveness via virus-induced changes in volatiles that affect aphid orientation, as tends to accompany nonpersistent virus infection (Mauck et al. 2010, 2012). I consider this less likely because aphids typically use volatile cues over short distances (i.e., several meters; Webster 2012), which should not influence landing rates at a broader field scale. Interestingly, early season colonizers landed more often in PRSV-infected than uninfected fields in 2010 suggesting a possible relationship, although they were only found in 4/11 infected fields and the evidence is relatively weak; they were virtually absent from fields in early 2011 as well. Regardless, it does not appear that the melon aphid, although a common colonizer, is an important vector of PRSV or WMV in pumpkins. I have repeatedly observed commercial pumpkin fields with severe viral epidemics, despite the growers reporting frequent insecticide use and thus maintaining ‘clean’ fields that are devoid of colonizing aphids. These results support the long-held opinion that nonpersistent viruses cannot be effectively managed with pesticides (Broadbent 1957). The negative relationship between noncolonizing aphids and WMV is harder to explain, but may be indirectly related with proximity to virus reservoirs. In 2010, bird cherry-oat aphid was the numerically dominant noncolonizing aphid, and its negative relationship with WMV prevalence remained strong both early in the season and overall. This species feeds on cereals and pasture grasses (Blackman & Eastop 1994), which are not hosts for WMV. In fact, sorghum (Sorghum bicolor L.), has even been used as an intercrop to control WMV in pumpkins (Damicone et al. 2007). Many of the study sites were surrounded by land containing cereals and pasture grasses, and the extent of coverage varied (unpublished data). If virus infection in pumpkin fields is related to the occurrence of virus reservoirs in the surrounding landscapes, it could follow that bird cherry-oat aphid is more abundant in fields surrounding cereals or pastureland, and therefore less likely to appear in fields with a high degree of infection due to the lack of nearby virus sources.

Species-Specific Vectors It may be common for nonpersistent virus epidemics to be driven by only a few noncolonizing species in both cucurbits (Adlerz 1978, Webb et al. 1994) and other crop systems as well (Raccah et al. 1985, Perez et al. 1995, Nebreda et al. 2004). For PRSV, the two aphid species associated with virus presence were cowpea aphid and spotted alfalfa aphid. The cowpea aphid is a known PRSV vector (Kumar et al. 2010), whereas spotted alfalfa aphid, to my knowledge, has not yet been tested using transmission assays. The early season alightment rate of cowpea aphids was greater in 2010 PRSV-infected fields as well, although counts were low and the relationship is weaker than that of overall alightment rates summed across the season. This could indicate that in addition to early-season specific factors such as high contrast between plant and soil facilitating cowpea aphid alightment in pumpkin fields, other season-long factors such as surrounding landscapes may play a role. Interestingly, both of these aphids are colonizers of leguminous plants such as alfalfa. Although it is unknown whether alfalfa is a compatible PRSV host, it can host WMV—a close relative to PRSV (Quiot-Douine et al. 1990). Future studies may benefit from testing this crop, and other leguminous plants, as a potential PRSV reservoir and the possibility that these two legume specialists transmit virus from their natal host plant. At least 29 aphid species can transmit WMV (Purcifull 1981), including the oleander aphid (Coudriet 1962, Yamamoto et al. 1982), which was the only species associated with WMV in my analyses. The oleander aphid prefers plants in the Apocynaceae family such as milkweed, a dominant field-edge-inhabiting plant in the Midwestern U.S. (Blackman and Eastop 1994Blackman and Eastop 1994). It is currently unknown whether milkweeds host WMV, thereby facilitating virus spread into pumpkin fields. Because of the near ubiquity of WMV among fields, it is possible that my community analysis did not detect other ecologically important links by which this prevalent virus infects fields. Conclusions Overall, my study identified the following key points. First, WMV is a dominant pumpkin virus of the Midwest. That being said, it is unclear how much this virus alone reduces crop yield. In the most heavily affected fields, I often detected co-infection by multiple viruses. Thus, cucurbits may be relatively tolerant of WMV as a single infection with concomitant reductions in plant growth, vigor, and reproduction when simultaneously infected by PRSV, CMV and/or ZYMV. These viral synergies warrant additional study. Second, broad categorizations (e.g., total aphids, colonizing/noncolonizing aphids), while convenient, are poor predictors of virus incidence. This is especially the case for colonizing aphids that are often

assumed to be the culprit due to their apparency. More nuanced approaches to pest management that entail species-specific targets would show far more promise in ameliorating the effects of viruses in these systems. Last, WMV and PRSV originated from extra-field sources, but little is yet known about the cultivated and wild reservoirs for these viruses. Given the idiosyncratic patterns regarding noncolonizing aphids, I anticipate that uncovering the infection status of natal host plants for key vectors in the greater landscape holds the most promise for breaking virus transmission cycles.

1.5 References

Adlerz, W. 1978. Secondary spread of watermelon mosaic virus 2 by Anuraphis middletonii. Journal of Economic Entomology 71: 531-533. Amemiya, T. 1985. Advanced Econometrics, Harvard University Press. Anderson, M. J. 2001. A new method for non‐parametric multivariate analysis of variance. Austral ecology 26: 32-46. Anderson, M.J. 2006. Distance-based tests for homogeneity of multivariate dispersions. Biometrics 62: 245-253. Atri, B.I., D.A. Enobakhare, and G. Thottappilly. 1986. The importance of colonizing and noncolonizing aphid vectors in the spread of cowpea aphid-borne mosaic virus in cowpea. Crop Protection 5: 406-410. Blackman, R. L., and V. F. Eastop. 1994. Aphids on the world's trees: an identification and information guide, CAB International, Wallingford. Borer, E.T., E.W. Seabloom, C.E. Mitchell, and A.G. Power. 2010. Local context drives infection of grasses by vector-borne generalist viruses. Ecology Letters 13: 810-818. Boydston, L. L., and D. W. Allison. 2003. Aphids of Istern North America north of Mexico: with keys to subfamilies and genera for female alatae, Washington State University Extension, College of Agricultural, Human and Natural Resource Sciences. Broadbent, L. 1957. Insecticidal control of the spread of plant viruses. Annual Review of Entomology 2: 339-354. Brust, G.E. 2000. Reflective and black mulch increase yield in pumpkins under virus disease pressure. Journal of Economic Entomology 93: 828-834. Brust, J.E., and K.L. Everts. 2010. Commerical vegetable production recommendations. EB 236. University of Maryland in cooperation with Rutgers the State University of New Jersey, Virginia Polytechnic Institute and State University, The Pennsylvania State Univesrity, and University of Delaware. Burnham, K. P., and D. R. Anderson. 2002. Model Selection and Multimodel Inference: A Practical Information-Theoretic Approach, vol. 2nd ed., Springer-Verlag. Clarke, K. R. 1993. Non‐parametric multivariate analyses of changes in community structure. Australian jmynal of ecology 18: 117-143.

Coudriet, D. L. 1962. Efficiency of various insects as vectors of cucumber mosaic and watermelon mosaic viruses in cantaloups. Jmynal of Economic Entomology 55: 519-520. Coutts, B., M.A. Kehoe, and R.A.C. Jones. 2012. Contact transmission of Papaya ringspot virus, Watermelon mosaic virus, and Squash mosaic virus in zucchini and lack of seed transmission in pumpkin and zucchini, pp. 87-90. In Horticulture Australia Ltd. Developing and communicating strategies for controlling virus diseases in vegetable cucurbit crops. Sydney, Australia. Coutts, B., M.A. Kehoe, and R.A.C. Jones. 2013. Zucchini yellow mosaic virus: Contact transmission, stability on surfaces, and inactivation with disinfectants. Plant Disease 97: 765-771. Damicone, J.P., J.V. Edelson, J.L. Sherwood, L.D. Myers, and J.E. Motes. 2007. Effects of border crops and intercrops on control of cucurbit virus diseases. Plant Disease 91: 509- 516. Davis, J.A. 2005. Soybean aphid, Aphis glycines Matsumura, a new vectors of Potato virus Y in potato. American Journal of Potato Research 82: 197-201. Davis, J.A., and E.B. Radcliffe. 2008. The importance of an invasive aphid species in vectoring a persistenly transmitted potato virus: Aphis glycines is a vector of potato leafroll virus. Plant Disease 92: 1515-1523. DiFonzo, C. D., D. W. Ragsdale, E. B. Radcliffe, N. C. Gudmestad, and G. A. Secor. 1997. Seasonal abundance of aphid vectors of Potato virus Y in the Red River Valley of Minnesota. Journal of Economic Entomology 90: 824-832. Douine, L., J.B. Quiot, G. Marchoux, and P. Archange. 1979. Recensement de espéces végetales sensibles au virus de la mosaïque du concombre (CMV). Étude bibliographique. Annual Review of Phytopathology 11: 439-475. Dusi, A.N., D. Peters, and W. Van Der Werf. 2000. Measuring and modeling the effects of inoculation date and aphid flights on the secondary spread of Beet mosaic virus in sugar beet. Annals of Applied Biology 136: 131-146. Egel, D. 2013. Pumpkin disease management timeline for Indiana: Purdue extension. Southwest Purdue Agricultural Center, Vincennes, IN. Favret, C., and D.J. Voegtlin. 2001. Migratory aphid (Hemiptera: Aphididae) habitat selection in agricultural and adjacent natural habitats. Environmental Entomology 30: 371-379.

Fereres, A.P., P. Pérez, C. Gemeno, and F. Ponz. 1993. Transmission of Spanish pepper- and potato-PVY isolates by aphid (Homoptera: Aphididae) vectors: Epidemiological implications. Environmental Entomology 22:1260-1265. Fletcher, J.D., A.R. Wallace, and B.T. Rogers. 2000. Potyviruses in New Zealand buttercup squash (Cucurbita maxima Dutch.): Yield and quality effects of ZYMV and WMV 2 virus infections. New Zealand Journal of Crop and Horticultural Science 28: 17-26. Gildow, F.E., D.A. Shah, W.M. Sackett, T. Butzler, B.A. Nault, and S.J. Fleischer. 2008. Transmission efficiency of Cucumber mosaic virus by aphids associated with virus epidemics in snap bean. Phytopathology 98: 1233-1241. Hanafi, A., E. Radcliffe, and D. Ragsdale. 1989. Spread and control of potato leafroll virus in Minnesota. Jmynal of economic entomology 82: 1201-1206. Hu, J., H. Li, K. Barry, M. Wang, and R. Jordan. 1995. Comparison of dot blot, ELISA, and RT-PCR assays for detection of two cucumber mosaic virus isolates infecting banana in Hawaii. Plant Disease 79: 902-906. Jossey, S., and M. Babadoost. 2008. Occurrence and distribution of pumpkin and squash viruses in Illinois. Plant Disease 92: 61-69. Kagezi, G.H. D.J. Voegtlin, and R.A. Weinzierl. 1999. The aphids (Homoptera: Aphididae) associated with bell peppers and surrounding vegetation in southern Illinois. The Great Lakes Entomologist 32: 161-173. Kanavaki, O. M., J. T. Margaritopoulos, N. I. Katis, P. Skmyas, and J. A. Tsitsipis. 2006. Transmission of Potato virus Y in tobacco plants by Myzus persicae nicotianae and M. persicae s.str. Plant Disease 90: 777-782. Katis, N. I., J. A. Tsitsipis, M. Stevens, and G. Powell. 2007. Transmission of Plant Viruses. In H. F. van Emden and R. Harrington (eds.), Aphids as Crop Pests. CABI, Cambridge, MA. Katis, N.I., J.A. Tsitsipis, D.P. Lykouressis, A. Papapanayotou, J.T. Margaritopoulos, G.M. Kokinis, D.C. Perdikis, and I.N. Manoussopoulos. 2006. Transmission of zucchini yellow mosaic virus by colonizing and noncolonizing aphids in Greece and new aphid species vectors of the virus. Journal of Phytopathology 154: 293-304. Kennedy, J. F., and E. S. Keeping. 1962. Linear regression and correlation, pp. 252-285. In V. Nostrand (ed.), Mathematics of Statistics, vol. 3rd ed., Princeton, NJ.

Kumar, N.K.K., H.S. Singh, and C.M. Kalleshwaraswamy. 2010. Aphid (Aphididae: Homopotera) vectors of papaya ringspot virus (PRSV), bionomics, transmission efficiency and factors contributing to epidemiology. Acta Horticulturae 951: 431-443. Larkin, M.A., G. Blackshields, N.P. Brown, R. Chenna, P.A. McGettigan, H. McWilliam, F. Valentin, I.M. Wallace, A. Wilm, and R. Lopez. 2007. Clustal W and Clustal X version 2.0. Bioinformatics 23: 2947-2948. Legendre, P. and L. Legendre. 1998. Numberical Ecology. 2nd 3d. Elsevier. Madden, L.V., T.P. Pirone, and B. Raccah. 1987. Analysis of spatial patterns of virus-diseased tobacco plants. Phytopathology 77: 1409-1417. Mauck, K.E., C.M. De Moraes, and M.C. Mescher. 2010. Deceptive chemical signals induced by a plant virus attract insect vectors to inferior hosts. Proceedings of the National Academy of Sciences USA 107: 3600-3606. Mauck, K.E., N.A. Bosque-Perez, S.D. Eigenbrode, C.M. De Moraes, and M.C. Mescher. 2012. Transmission mechanisms shape pathogen effects on host-vector interactions: Evidence from plant viruses. Functional Ecology 26: 1162-1175. Miller, A. 2002. Subset Selection in Regression, 2nd ed., vol. 95, Chapman & Hall/CRC. Mora-Aguilera, G., D. Nieto-Angel, C.L. Campbel, D. Téliz, E. García. 1996. Multivariate comparison of papay ringspot epidemics. Phytopathology 86: 70-78. Mueller, E.E., K.E. Frost, P.D. Esker, and C. Gratton. 2010. Seasonal phenology of Aphis glycines (Hemiptera: Aphididae) and other aphid species in cultivated bean and noncrop habitats in Wisconsin. Journal of Economic Etnomology 103: 1670-1681. Nault, L.R. 1997. Arthropod transmission of plant viruses: a new synthesis. Annals of the Entomological Society of America 90: 521-541. Nault, B.A., D.A. Shah, K.E. Straight, A.C. Bachmann, W.M. Sackett, H.R. Dillard, S.J. Fleischer, and F.E. Gildow. 2009. Modeling temporal trends in aphid vector disdpersal and cucumber mosaic virus epidemics in snap beans. Environmental Entomology 38: 1347-1359. Nebreda, M., A. Moreno, N. Perez, I. Palacios, V. Seco-Fernandez, and A. Fereres. 2004. Activity of aphids associated with lettuce and broccoli in Spain and their efficiency as vectors of lettuce mosaic virus. Virus Research 100: 83-89. Ng, J.C.K. and B.W. Falk. 2006. Virus-vector interactions mediating nonpersistent and semipersistent transmission of plant viruses. Annual Review of Phytopathology 44: 183- 212.

Paulsrud, B. 2005. Workshop summary, p. 109. In Midwest Pest Management Strategic Plan for Processing & Jack-o-Lantern Pumpkins, Illinois, Indiana, Iowa and Missouri. University of Illinois Extension, Urbana, IL. Perez, P., J.L. Collar, C. Avilla, M. Duque, and A. Fereres. 1995. Estimation of vector propensity of potato virus Y in open-field pepper crops of central Spain. Journal of Economic Entomology 88: 986-991. Perring, T.M., R.N. Royalty, and C.A. Farra. 1989. Floating row covers for the exclusion of virus vectors and the effect on disease incidence and yield of cantaloupe. Journal of Economic Entomology 82: 1709-1715. Peters, D., E. Booijmans, and P. F. M. Gronhuis. 1990. Mobility as a factor in the efficiency with which aphids can spread non-persistently transmitted viruses. Proceedings of the Section Experimental and Applied Entomology of the Netherlands Entomological Society 1: 190-194. Purcifull, D. 1981. Watermelon mosaic 2 potyvirus. In A. A. Brunt, K. Crabtree, M. J. Dallwitz, A. J. GWebbs, L. Watson and E. J. Zurcher [eds.], Plant Viruses Online: Descriptions Lists from the VIDE Database. Version: 20th August 1996. Quiot-Douine, L., H. Lecoq, J. Quiot, M. Pitrat, and G. Labonne. 1990. Serological and biological variability of virus isolates related to strains of papaya ringspot virus. Phytopathology 80: 256-263. Raccah, B., A Gal-On, and V. Eastop. 1985. The role of flying aphid vectors in the transmission of cucumber mosaic virus and potato virus Y to peppers in Isreal. Annals of Applied Biology 106: 451-460. R Core Team. 2011. R: A Language and Environment for Statistical Computing. computer program, version 2.15.2. By, Vienna, Austria. Seabloom, E.W., P.R. Hosseini, A.G. Power, and E.T. Borer. 2009. Diversity and composition of virus communities: Coinfection of barley and cereal yellow dwarf viruses in California grasslands. American Naturalist 173: E79-E98. Smith, C. F., R. W. Eckel, and E. P. Lampert. 1992. A key to many of the common alate aphids of North Carolina (Aphididae: Homoptera). Technical bulletin. Summers, C.G. J.R. Newton, M. Kirk, and S.R. Temple. 1990. Transmission of beet yellow and beet mosaic viruses by noncolonizing aphid vectors. Journal of Economic Entomology 83: 2448-2451.

Walters, S.A. 2003. Viruses associated with cucurbit production in southern Illinois. HortScience 38: 65-66. Wakman, W., M.S. Kontonga, D.S. Teakleb, and D.M. Persley. 2002. Watermelon mosaic virus of pumpkin (Cucurbita maxima) from Sulawesi: identification, transmission, and host range. Indonesian Journal of Agricultural Science 3: 33-36. Webb, S.E., M.L. Kok-Yokomi, and D.J. Voegtlin. 1994. Effect of trap color on species composition of alate aphids (Homoptera: Aphididae) caught over watermelon plants. Florida Entomologist 77: 146-154. Webster, B. 2012. The role of olfaction in aphid host location. Physiological Entomology 37: 10- 18. Yamamoto, T. M., T. Ishii, T. Katsube, and M. Sorin. 1982. Transmission of watermelon mosaic virus by aphid species (Hemiptera: Aphididae). Japanese Jmynal of Applied Entomology and Zoology 26: 218-223. Zitter, T. A., D. L. Hopkins, and C. E. Thomas (eds.). 1996. Compendium of Cucurbit Diseases. The American Phytopathological Society, St. Paul, MN.

CHAPTER 2. COMPARING LANDCAPE SCALES OF INFLUENCE IN CROP VIRUS EPIDEMIOLOGY

2.1 Introduction

Landscape-level interactions remain starkly under-researched in crop-disease ecology, and multiple calls have recently been made to prioritize identification of land cover and spatial scale interactions and better inform prevention and management practices (Alexander et al. 2014, Cunniffe et al. 2015). While there is evidence for within-field (e.g., Ali et al. 2012, Duffus 1971, Smith et al. 2012, Srinivasan et al. 2013) and surrounding land cover (e.g., Carrière et al. 2014, Fabre et al. 2005, Margosian et al. 2009) mediation of insect-vectored crop virus, to our knowledge there have been no comparisons of the relative strength of influence. Further, there is substantial potential for the mediation of stylet-borne, or nonpersistent virus epidemiology by surrounding land cover, but this too remains largely uninvestigated (but see Mueller et al. 2012). With those literature gaps in mind, we explore land cover interactions in a nonpersistent crop virus system, and implement a direct comparison of within-field versus surrounding landscape influence on virus epidemiology. There are several ways in which nonpersistent virus epidemiology may be mediated by interaction with surrounding land or within-field weed cover. Aphid-vectored nonpersistent plant viruses often have large numbers of potential vectors, including noncolonizing species (i.e., those that do not feed and reproduce on a plant). In fact, several studies have documented the relative importance of noncolonizing aphid species’ transient alightment within fields over that of colonizing species in virus epidemiology (e.g., Raccah et al. 1985, Summers et al. 1990, Fereres et al. 1992, 1993, Webb et al. 1994, Perez et al. 1995, Nebreda et al. 2004). Vector source populations can thereby be traced back to many different habitats, and such a broad range of vectors results in likewise disparate and wide-ranging mediators of vector community and viral infection dynamics. Additionally complicating matters, nonpersistent viruses tend to have very large plant-host ranges: for example, over 800 species of plants can host cucumber mosaic virus (Zitter et al. 1996). Thus, both vector and virus reservoirs can lurk in the surrounding landscape or within fields, affecting the probability of virus infection within crop systems. Finally, the

landscape can also mediate virus infection via aphid response to visual and structural variation via windbreaks, and attraction to color and/or contrast (e.g., Mayse & Price 1978, Moericke 1955, Bottenberg & Irwin 1992, respectively), although behavioral responses to visual cues can be species-specific. Numerous variables embedded in the landscape can influence stylet-borne, or nonpersistent, plant virus epidemiology, but the scale at which these variables have the largest mediating influence, or the scale of effect, in vector-virus dynamics is unknown. Scale of effect is strongly linked to dispersal distance (Jackson & Fahrig, 2012), and there is a great range in recorded dispersal distances among aphids. At sufficiently low wind speeds aphids are capable of sustaining short intervals of directed flight for an estimated maximum 200 m, losing control in wind speeds above ca. 0.6 m s-1 (Haine 1955, Loxdale et al. 1993, Parry 2013). In the state of Indiana, prevailing winds average 4.3 m s-1 from the Southwest (ICLIMATE.ORG 2015), aphid dispersal may often be assisted by the wind. In contrast, long-distance migratory flights occurring via passive wind dispersal commonly cover ca. 20–50 km, but are thought to be much less frequent among aphid populations than localized movement (Loxdale et al. 1993). Pumpkins are host to several aphid-vectored nonpersistent viruses (Zitter et al. 1996), and the aphid species and viruses implicated in this system were recently described (Angelella et al. 2015). We attempt to identify the scale of effect for vectors and corresponding nonpersistent virus infection in focal pumpkin fields, and the interactions between virus reservoirs and vectors in the landscape. We used the approach of quantifying land cover components within concentric buffer zones surrounding focal sites (Brennan et al. 2002), and finding scales best predicting vector behavior and virus infection (Jackson & Fahrig 2012), exploring the interactions that occur therein. 2.2 Methods Aphid, Virus Quantification Aphid species alightment and virus infection were quantified within Indiana pumpkin fields in 2010 (n=10) and 2011 (n=16) (see Angelella et al. 2015). Although pumpkins host four aphid-vectored nonpersistent viruses–watermelon mosaic virus type 2 (WMV; Potyviridae), papaya ringspot virus type W (PRSV; Potyviridae), zucchini yellow mosaic virus (ZYMV; Potyviridae), and cucumber mosaic virus (CMV; Bromoviridae) (Zitter et al. 1996)–only WMV and PRSV were detected in pumpkin leaf assays collected at the end of the growing season. Fields were under varied management regimes.

Cover Quantification Weed species were surveyed during the growing season of 2010 and 2011. Specimens were surveyed twice in 2010 (late July and late August) by walking ten steps on a random trajectory at three separate locations into each field, and collecting all weeds within a two meter radius. In 2011, a survey of weed species occurred every three weeks from mid-June through the end of September within each field. Stopping at four occurrences each along four transects between randomly-selected rows at 20 m intervals, weeds within a 0.75 m x 0.75 m quadrat were identified. Weeds were identified to species except those in the genera Amaranthus, Prunus, or Trifolium, and grasses were identified to species or genus when possible (Newcomb 1977, Uva et al. 1997). Cover was quantified as number of stems species-1 and total number of stems in 2010, weed cover m-2 species-1 and total weed cover m-2 in each field were generated from 2011. Any aphids found on weeds during surveys were collected in 70% ethanol for subsequent identification; however, only a single specimen was found (Aphis nerii Boyer de Fonscolombe on eastern black nightshade, Solanum ptycanthum Dunn). To obtain surrounding landscape composition, GPS coordinates were found for each field. GIS-based analyses of National Land Cover Data satellite images were conducted to generate land cover values in a 5 km radius, and grouped within successive 1-km intervals (i.e., 1–5 km). Landscape data were collected from the National Agricultural Statistics Service (NASS) (2011). Cover data were converted to proportion of the surrounding landscape at each spatial scale from pixels, with each pixel representing a 900 m2 area, and one of ca. 80 land cover classes (MRLC 2007). Variable Selection Aphid variables included in analyses were alightment totals of all noncolonizers (all aphids, excluding Aphis gossypii Glover). Aphis gossypii was observed colonizing and reproducing on pumpkins within fields; therefore, alightment totals from individuals flying into fields were confounded by the number of individuals originating within fields (see Angelella et al. 2015). Because the aim of this study is to explore effects of landscape features at different spatial scales on vector alightment within fields and subsequent virus infection, A. gossypii was not included. To standardize 2010 aphid alightment data, totals were divided by the number of data collection weeks. All 2011 aphid alightment totals were generated from the same sampling time-frame, and are analyzed in raw form. Land cover types were combined when biologically relevant, and retained if >10% of the landscape (Table 2-1). Variables included land characterized by row crops, woody vegetation,

pasture, grass or hay fields, and 20–100% constructed materials such as buildings or asphalt. These variables were selected to encompass sufficient potential aphid natal host habitat and structural diversity for the observation of reservoir or behavioral mediation, in large enough proportions as to be distinct from environmental stochasticity. Inventoried weed species susceptible to WMV include the following: Amaranthus spp., Chenopodium album L., Datura stramonium L., Digitaria sanguinalis (L.), Ipomoea hederacea Jacq., Mollugo verticillata L., Phytolacca americana L., Solanum ptycanthum, and Trifolium spp. (Angelella unpublished, Brunt et al. 1996, Zitter 2001). Only one PRSV-susceptible species was present in cover surveys: Chenopodium album (Brunt et al. 1996).

Table 2-1. Land cover variables included in analyses, and cover composition by National Agricultural Statistics Service (NASS) category. Land Cover Variables NASS Categories Included Row Crops Corn, Sweet Corn, Popcorn, Winter Wheat, Soybeans Pasture/Hay/Grassland Alfalfa, Pasture Hay, Other Hay Non-Alfalfa, Sod Grass Seed, Grassland, Developed Open Space Forest/Shrubland Deciduous Forest, Evergreen Forest, Mixed Forest, Woody Wetlands, Shrubland Urban Developed Low Intensity, Developed Medium Intensity, Developed High Intensity

Prior to analyses, the proportions of pumpkin leaf samples infected with WMV or PRSV in each field were arcsine square root transformed. Land cover data were standardized within each spatial scale (1–5 km). An outlier was removed from 2011 total alightment data, to prevent it from driving significant results. Analyses were conducted in R version 2.12.1 software (R Core Team 2012). Data Analyses To compare aphid alightment and PRSV or WMV infection variation within fields by landscape cover at each spatial scale and within-field weed cover, partial least squares path model (PLS-PM) analyses were conducted using the “semPLS” package (Monecke & Leisch 2012). There are several situations in which PLS-PM is especially useful, including when conducting exploratory analyses, building complex models, utilizing small sample sizes, and using formative indicator variables (Chin 2010). The PLS-PM analyzes linear relationships between latent variables using multiple ordinary least squares regressions, in which each latent variable summarizes blocks of observed and quantified manifest variables (Sanchez 2013). Latent variables are considered reflective or formative depending on their relationship to manifest variables. If manifest variables measure the same underlying concept and changes in one implies

change in all of the variables in a block, they are considered reflective; formative variables may be uncorrelated, on the other hand, and do not necessarily measure the same underlying concept (Sanchez 2013). In our models, the landscape at each spatial scale surrounding pumpkin fields [“land cover (1 km)”, “land cover (2 km)”, “land cover (3 km)”, “land cover (4 km)”, or “land cover (5 km)”] was considered a formative latent variable, with land cover variables “row crops”, “forest/shrubland”, “pasture/hay/grass”, and “urban”. The remaining latent variables were measured with a single manifest variable and considered reflective: the aphid alightment variable is comprised of aphid alightment totals excluding Aphis gossypii; both “PRSV” and “WMV” occurrence were measured by the proportion of PRSV- or WMV-infected pumpkin tissue samples per field, respectively; and latent variable “weed cover” was measured by either total stems (2010), or total cover m-2 field-1 (2011). Because all of the latent variables with multiple contributing measures are formative and there are no prior informative models, bootstrap resampling was used to validate the predictive abilities of the path models (Chin 2010). Specifically, bootstrap resampling evaluated the significance of direct path coefficients and the weights of manifest variables within formative blocks. Path model graphics were generated with Graphviz – Graph Visualization Software (Gansner & North 2000). Lastly, individual cover effects on vector alightment and PRSV or WMV infection were assessed with Poisson or linear regressions, respectively. These include assessing reservoir weed and crop species as virus predictors, assessing total weed cover as virus or aphid alightment predictors, and assessing surrounding cover types within the best predicting land cover scales as alightment predictors. Regressions were also performed to assess the contribution of land cover types as virus reservoirs on virus occurrence within fields, and interactive effects between aphid alightment and reservoir species on virus occurrence. 2.3 Results Cover as Mediator of Vector Dispersal For the sake of brevity, PLS-path models are shown including weed cover and the single significant surrounding landscape scale for each year (Fig. 2-1). Full summaries of outer model (Table B2-1) and inner model (Table B2-2) coefficients and bootstrap values are available in the Appendix. All manifest variables fall within the range of bootstrap-predicted values, suggesting a fairly good overall predictive accuracy of the model.

B) Figure 2-1. PLS-path models of total aphid alightment excluding Aphis gossypii within pumpkin fields and predicted relationship with surrounding land cover, within-field weed cover, and virus infection in crops. A) 2010, B) 2011

Surrounding land cover is a consistently stronger predictor of aphid alightment than within-field weed cover in models. Weed cover is a significant but relatively smaller contributor to aphid alightment in the 2010 PLS-path model, but is not a significant contributor to aphid alightment in 2011. The scale of the strongest predicting surrounding land cover was consistent across years at 3 km. The largest-weighted cover types were row crops in both years Individual regressions revealed a significant negative association between weed cover and aphid alightment in 2010 (Fig. 2-2), and several surrounding land cover types significantly predicted increases or decreases in aphid alightment in 2011 (Fig. 2-3). The proportion of row crops in the surrounding landscape within 3 km was positively associated with aphid alightment, while pasture/hay/grass and forest/shrubland were negatively associated.

Figure 2-2. Poisson regression of mean total aphid alightment week-1 field-1 by total within-field weed cover field-1 (total stem counts) within pumpkin fields in 2010 (n=10).

B) C)

Figure 2-3. Poisson regression of total aphid alightment field-1 by surrounding land cover (proportion of cover) within 3 km of 2011 pumpkin fields (n=16): A) row crops, B) pasture/hay/grass, C) forest/shrubland.

Cover as Virus Reservoir No PRSV- or WMV-reservoir weed species significantly predicted virus occurrence (Table B2-3). Additionally, no significant interactions occurred between total aphid alates and virus reservoirs to predict either PRSV or WMV infection (Tables B2-4, B2-5). 2.4 Discussion Results reflected in the PLS-path models suggest cover within the surrounding landscapes exerts a greater influence on vector alightment than cover within fields. An increase in within-field weed cover predicts a corresponding decrease in aphid alightment in 2010 (Fig. 2-2), but the predictive ability as indicated by path model coefficient is small relative to that of the surrounding landscape (Fig. 2-1). Within-field weed cover is not a significant predictor of aphid alightment variation in 2011. In both years, the surrounding landscape within 3 km significantly predicts aphid alightment; the 3 km scale is thus the most influential in modifying overall within the focal fields. In particular, the “row crops” variable is weighted heavily, and predicts an increase in alightment with increased coverage in the surrounding landscape in 2011. “Forest/shrubland” and “pasture/hay/grass” are negatively associated with aphid alightment in 2011. The positive relationship between the proportion of row crops in the surrounding landscape and aphid alightment could indicate their function as natal habitats. Some of the most abundant species’ host ranges include row crops. The bird cherry-oat aphid (Rhopalosiphum padi

L.) prefers poaceous hosts such as corn and wheat, the corn leaf aphid [Rhopalosiphum maidis (Fitch)] also colonizes corn, and the cowpea aphid (Aphis craccivora Koch) can colonize soybeans (Blackman 2015). Although the cowpea aphid and bird cherry-oat aphid inhabit crops in the “pasture/hay/grass” category such as alfalfa and Timothy-grass, in increase in proportion of this cover type predicted a decline in aphid alightment. Relationships between aphid alightment and surrounding land cover may also reflect their structural effects on the dispersal behavior of aphids in flight, such as contrast or wind-breaks. The negative impact of forest/shrubland on aphid counts may reflect a wind-break effect blocking alightment within fields. The effect of a wind break on aphid alightment would likely depend on wind direction relative to the field. Greater concentrations of wind-dispersed herbivores are found leeward of windbreaks (Mayse & Price 1978), and aphids prefer landing with their bodies oriented into the wind (Storer et al. 1999), which may explain why aphids are often found near windbreaks rather than in them. However, various types of visual stimuli may be more important predictors of alightment than windbreaks (e.g., Bottenberg & Irwin 1992, Favret & Voegtlin 2001). Upon leaving migratory flight to engage in host-plant search behaviors, aphids display a shift in light intensity preference to green and yellow wavelengths (500 – 600 nm) such as would be reflected by foliage (Moericke 1955, Kennedy et al. 1961, Nottingham et al. 1991). Further, most aphid species exhibit a preference for alighting upon green/yellow targets surrounded by contrasting backgrounds (Kennedy et al. 1961), and ground cover appears to be one of the most important factors in large-scale plant architecture affecting rates of aphid alightment (Bottenberg & Irwin 1992, Ogenga -Latigo et al. 1992, Döring et al. 2004, Hooks & Fereres 2006, George et al. 2012), more so than barrier height or barrier density. This preferential orientation toward high or low degrees of ground cover and contrast is likely species-specific (e.g., Favret & Voegtlin 2001). The positive association of row crops observed on alightment may reflect a preference for contrast serving to collect alates in proximity to pumpkin fields. Aphid alightment was also negatively predicted by weed cover within fields, further suggesting a preference for contrast. A previous study likewise found a greater number of total aphids in the air above open-canopy crops relative to a closed-canopy prairie (Favret & Voegtlin 2001). Aphid overall scales-of-effect, indicated by the best significant predictive power by land cover scale, were consistent between years. In both models the best-predicting scale fell in the middle of the distance continuum, indicating variables were quantified within an optimal distance gradient (Jackson & Fahrig 2015). The consistency is somewhat surprising because dispersal capabilities among species vary widely, as is well-documented by flight mill studies [e.g., R.

padi, Cheng et al. 1997; Rhopalosiphum maidis (Fitch), Bottenberg & Irwin 1992; Aphis fabae (Scop.), Nottingham & Hardie 1989; Sitobion avenae (F.), Cheng et al. 2002; Aphis gossypii, Liu et al. 2003]. Species may be considered migratory or non-migratory based on these abilities, with migratory aphids predictably engaging in longer flights (Kring 1972). Additionally, certain morphologies are primed for longer flights than others. The duration of flights initiated by autumn-flying gynoparae and males is over three times the duration of summer-flying alate virginoparae, and half again that of spring-flying fundatrigeniae morphs of A. fabae (David & Hardie 1988, Nottingham et al. 1991). However, Loxdale and Hardie (1993) argued that the shorter flights characteristic of summer morphs are likely more biologically relevant than longer flights concerning population or genotype distributions and virus epidemiology. Shorter flights occur with greater frequency than longer migratory flights, and many economically important crops are vulnerable to virus damage during the spring and summer after migratory flights occur. A preponderance of shorter flights by these summer morphs would thus expected to pull median aphid dispersal distances closer to focal pumpkin fields, as our results suggest. Pooling aphid alightment data across species may have an averaging effect as well. Comparisons of virus reservoir influence by proximity to fields cannot be made, because evidence of reservoir effects were absent within the surrounding landscape and within field weed communities. Within-field coverage of weedy virus reservoir species did not predict virus infection, and neither soybean nor alfalfa coverage in the surrounding landscape predicted WMV infection. There are limited data documenting PRSV susceptibility among plant species, and it is likely that reservoirs within field weed communities (or surrounding landscapes) have not yet been identified. There was a very low infection detection frequency among weeds growing within fields, however. An assay of weed samples collected from the study field sites detected WMV at a very low rate and did not detect PRSV at all (Angelella unpublished). One would expect higher rates of infection among susceptible perennials in the surrounding landscape such as alfalfa, leading to a “spillover effect” of high infection rates within nearby crops (e.g., Seabloom et al. 2009, Borer et al. 2010), but we found no support for this. It is possible that our snapshot of cumulative virus and vector occurrence within fields lacks the resolution needed to pinpoint virus reservoir relationships. Conclusions Evidence suggesting mediation of aphid dispersal by both within-field and surrounding land cover was found, though the predictive ability of surrounding land cover was consistently stronger than that of within-field weed cover. In both years, surrounding land cover within the 3

km scale had the strongest impact on aphid alightment within fields. Structural variation providing contrast cues in row crops and wind-break effects in forest/shrubland may have attracted or deterred alightment in the vicinity of fields, respectively, and row crops could have provided aphid natal source habitats as well. Lastly, no relationships among virus reservoirs and WMV or PRSV infection were found. Studies with greater temporal resolution may increase likelihood of detecting a reservoir-infection link. Results have implications for nonpersistent virus prevention, suggesting weed management may not be an effective method.

2.5 References

Ali, A., O. Mohammad, and A. Khattab. 2012. Distribution of viruses infecting cucurbit crops and isolation of potential new virus-like sequences from weeds in Oklahoma. Plant Disease 96: 243-248. Brunt, A.A., K. Crabree, M.J. Dallwitz, A.J. GWebbs, L. Watson, and E.J. Zurcher (eds.). 1996. Plant Viruses Online: Descriptions and Lists from the VIDE Database. Version: 20th August 1996. Blackman, R. L. Aphids on the World's Plants. Natural History Museum, London. Borer, E. T., E. W. Seabloom, C. E. Mitchell, and A. G. PoIr. 2010. Local context drives infection of grasses by vector‐borne generalist viruses. Ecology letters 13: 810-818. Bottenberg, H., and M. E. Irwin. 1992. Flight and landing activity of Rhopalosiphum maidis (Homoptera: Aphididae) in bean monocultures and bean-corn mixtures. Journal of Entomological Science 27: 143-153. Brennan, J. M., D. J. Bender, T. A. Contreras, and L. Fahrig. 2002. Focal patch landscape studies for wildlife management: optimizing sampling effort across scales, pp. 68-91. In J. Liu, and W.W. Taylor (eds.). Integrating Landscape Eecology into Natural Resource Management. Carrière, Y., B. Degain, K. Hartfield, K. Nolte, S. Marsh, C. Ellers-Kirk, W. Van LeeuIn, L. Liesner, P. Dutilleul, and J. Palumbo. 2014. Assessing Transmission of Crop Diseases by Insect Vectors in a Landscape Context. Jmynal of Economic Entomology 107: 1-10. Chin, W. W. 2010. How to Write Up and Report PLS Analyses, pp. 655-690, Handbook of partial least squares. Springer. Cunniffe, N. J., B. Koskella, C. J. E. Metcalf, S. Parnell, T. R. Gottwald, and C. A. Gilligan. 2015. Thirteen challenges in modelling plant diseases. Epidemics 10: 6-10. David, C., and J. Hardie. 1988. The visual responses of free-flying summer and autumn forms of the black bean aphid, Aphis fabae, in an automated flight chamber. Physiological Entomology 13: 277-284. Duffus, J. E. 1971. Role of weeds in the incidence of virus diseases. Annual Review of Phytopathology 9: 319-340.

Fabre, F., M. Plantegenest, L. Mieuzet, C. A. Dedryver, J.-L. Leterrier, and E. Jacquot. 2005. Effects of climate and land use on the occurrence of viruliferous aphids and the epidemiology of barley yellow dwarf disease. Agriculture, ecosystems & environment 106: 49-55. Favret, C., and D. J. Voegtlin. 2001. Migratory aphid (Hemiptera: Aphididae) habitat selection in agricultural and adjacent natural habitats. Environmental Entomology 30: 371-379. Fereres, A., M. J. Blua, and T. M. Perring. 1992. Retention and transmission characteristics of zucchini yellow mosaic virus by Aphis gossypii and Myzus persicae (Homoptera: Aphididae). Jmynal of Economic Entomology 85: 759-765. Fereres, A., P. Perez, C. Gemeno, and F. Ponz. 1993. Transmission of Spanish pepper- and potato-PVY isoolates by aphid (Homoptera: Aphididae) vectors: Epidemiological implications. Environmental Entomology 22: 1260-1265. Gansner, E. R., and S. C. North. 2000. An open graph visualization system and its applications to software engineering. Software – Practice and Experience 30: 1203-1233. George, D. R., R. H. Collier, and D. M. Whitehouse. 2012. Can imitation companion planting interfere with host selection by Brassica pest insects? Agricultural and Forest Entomology 15: 106-109. Haine, E. 1955. The flight activity of the sycamore aphid, Drepanosiphum platanoides Schr.(Hemiptera, Aphididae). The Jmynal of Animal Ecology: 388-394. Hooks, C. R. R., and A. Fereres. 2006. Protecting crops from non-persistently aphid-transmitted viruses: a review on the use of barrier plants as a management tool. Virus Research 120: 1-16. ICLIMATE.org. 2015. Indiana State Climate Office. Dpt. Agronomy, Purdue University. Jackson, H. B., and L. Fahrig. 2012. What size is a biologically relevant landscape? Landscape ecology 27: 929-941. Jackson, H. B., and L. Fahrig. 2015. Are ecologists conducting research at the optimal scale? Global Ecology and Biogeography 24: 52-63. Karl, E., and K. Schmelzer. 1971. Investigations on the transmissibility of watermelon mosaic viruses by aphid species. Archiv für PflSchutz 7: 3-11. Kennedy, J. S., C. O. Booth, and W. J. S. Kershaw. 1961. Host finding by aphids in the field. III. Visual attraction. Annals of Applied Biology 49: 1-21. Kring, J. B. 1972. Flight behavior of aphids. Annual Review of Entomology 17: 461-492.

Loxdale, H. D., J. Hardie, S. Halbert, R. Foottit, N. A. C. Kidd, and C. I. Carter. 1993. The relative importance of short- and long-range movement of flying aphids. Biological Reviews 68: 291-311. Margosian, M. L., K. A. Garrett, and J. S. Hutchinson. 2009. Connectivity of the American agricultural landscape: assessing the national risk of crop pest and disease spread. BioScience 59: 141-151. Moericke, V. 1955. Uber die Lebensgewohnheiten der geflugelten Blattlause (under besonderer Berucksichtigung des Verhaltens beim Landen). Zeitschrift fur Angewandte Entomologie 37: 29-91. Monecke, A., and F. Leisch. 2012. semPLS: structural equation modeling using partial least squares. Journal of Statistical Software 48: 1-32. (MRLC) Multi-Resolution Land Characeteristics Consortium. 2011. NLCD 2001 Land Cover Class Definitions. U.S. Environmental Protection Agency. Mueller, E., R. Groves, and C. Gratton. 2012. Crop and non-crop plants as potential reservoir hosts of Alfalfa mosaic virus and Cucumber mosaic virus for spread to commercial snap bean. Plant Disease 96: 506-514. (NASS) National Agricultural Statistics Service. 2010. Cropland Data Layer. Washington D.C.: U.S. Department of Agriculture. http://www.nass.usda.gov/research/Cropland/SARS1a.htm Nebreda, M., A. Moreno, N. Perez, I. Palacios, V. Seco-Fernandez, and A. Fereres. 2004. Activity of aphids associated with lettuce and broccoli in Spain and their efficiency as vectors of Lettuce mosaic virus. Virus Research 100: 83-89. Newcomb, L. 1977. Newcomb's wildfloIr guide, Little, Brown, and Co., New York, NY. Ogenga-Latigo, M., J. Ampofo, and C. Baliddawa. 1992. Influence of maize row spacing on infestation and damage of intercropped beans by the bean aphid (Aphis fabae Scop.). I. Incidence of aphids. Field Crops Research 30: 111-121. Parry, H. R. 2013. Cereal aphid movement: general principles and simulation modelling. Movement Ecology 1: 1-15. Perez, P., J. L. Collar, C. Avilla, M. Duque, and A. Fereres. 1995. Estimation of vector propensity of potato virus Y in open-field pepper crops of central Spain. Jmynal of Economic Entomology 88: 986-991. R Core Team. 2011. R: A Language and Environment for Statistical Computing. computer program, version 2.15.2. By, Vienna, Austria.

Raccah, B., A. Gal‐On, and V. Eastop. 1985. The role of flying aphid vectors in the transmission of cucumber mosaic virus and potato virus Y to peppers in Israel. Annals of applied Biology 106: 451-460. Sanchez, G. 2013. PLS Path Modeling with R. Trowchez Editions, Berkeley. Seabloom, E. W., P. R. Hosseini, A. G. PoIr, and E. T. Borer. 2009. Diversity and composition of viral communities: coinfection of barley and cereal yellow dwarf viruses in California grasslands. The American Naturalist 173: E79-E98. Smith, E. A., A. DiTommaso, M. Fuchs, A. M. Shelton, and B. A. Nault. 2012. Abundance of weed hosts as potential sources of onion and potato viruses in Istern New York. Crop Protection 37: 91-96. Srinivasan, R., J. M. Alvarez, and F. Cervantes. 2013. The effect of an alternate weed host, hairy nightshade, Solanum sarrachoides (Sendtner) on green peach aphid distribution and Potato leafroll virus incidence in potato fields of the Pacific NorthIst. Crop Protection 46: 52-56. Storer, J. R., S. Young, and J. Hardie. 1999. Three-dimensional analysis of aphid landing behavior in the laboratory and field. Physiological Entomology 24: 271-277. Summers, E., J. Newton, M. Kirk, and S. Temple. 1990. Transmission of beet yellows and beet mosaic viruses by noncolonizing aphid vectors. Jmynal of economic entomology 83: 2448-2451. Uva, R. H., J. C. Neal, and J. M. DiTomaso. 1997. weeds of the Northeast, Comstock Pub. Associates, Ithaca, NY. Webb, S. E., M. L. Kok-Yokomi, and D. J. Voegtlin. 1994. Effect of trap color on species composition of alate aphids (Homoptera: Aphididae) caught over watermelon plants. Florida Entomologist 77: 146-154. Zitter, T. A., D. L. Hopkins, and C. E. Thomas (eds.). 1996. Compendium of Cucurbit Diseases. The American Phytopathological Society, St. Paul, MN.

CHAPTER 3. DISPERSAL DISTANCE AND HOST-PLANT ASSOCIATED PATTERNS OF COLONIZATION IN A NONPERSISTENT VIRUS VECTOR

3.1 Introduction

The ecological success of aphids is reflected in their ability to exploit ca. 300 plant families (Blackman 2015), and in the resultant status of many as important crop pests (Blackman & Eastop 1984). Yield reductions caused by aphid-vectored viruses in Midwestern pumpkins illustrate this point (Paulsrud 2005). Due to their small size and complex flight patterns, however, it is largely unknown how aphids disperse from source populations to arrive in focal crop fields. This study uses molecular markers to explore the population genetic structure of a putative virus vector in pumpkins, the cowpea aphid (Aphis craccivora Koch), and whether patterns in geographical distance or host-association relate to vector occurrence in focal fields. Dispersal ability among aphid species is quite variable, and prior studies display this in the varied relationships of geographic scale and aphid population genetic structure. Migratory species may exhibit genetic differentiation on a much larger spatial scale. For example, Aphis glycines Matsumura populations are positively correlated within a distance class as large as 300 km or less (Orantes et al. 2012), and Rhopalosiphum padi (L.), with its notable dispersal abilities, exhibited population differentiation only between populations >1000 km apart (Delmotte et al. 2002). Myzus persicae (Sulzer) populations differentiated at the smaller scale of 150–200 km in one study (Guillemaud et al. 2003), and another study suggested M. persicae dispersal can be much more limited by revealing genetic differentiation even in populations <50 km apart (Wilson et al. 2002). To my knowledge, the dispersal capabilities of the cowpea aphid have not yet been quantified. However, genetic differentiation among populations of obligate parthenogenetic aphids can be higher than that of cyclically parthenogenetic aphids (e.g., Delmotte et al. 2002, Guillemaud et al. 2003). The cowpea aphid is suspected to be obligate parthenogenetic across most of its range in North America (Blackman & Eastop 2008), and might thus exhibit relatively frequent differentiation at a smaller geographic scale.

Cowpea aphids are specialists of legumes, although they are known to colonize other plant families as well (Blackman 2015). Alfalfa (Medicago sativa L.) and black locust (Robinia pseudoacacia L.) are both preferred hosts of the cowpea aphid and commonly occur in the Midwest, making them ideal for scouting and sampling cowpea aphid populations. The relative contribution of these potential sources to cowpea aphid occurrence in pumpkin fields could have implications for pathogen management. Recently, the existence of two endosymbiont-mediated biotypes, ‘locust-associated-biotype’ and ‘alfalfa-associated-biotype’, were discovered in cowpea aphids (Wagner et al. 2015). It is therefore likely that differentiated genotypes occur between alfalfa- and black locust-associated populations. Prior population genetic studies have also successfully used molecular markers to determine whether host-associated genotypes exist in an aphid species (e.g., Sunnucks et al. 1997, Fuller et al. 1999, Komazaki et al. 2011, Sandrock et al. 2011). Should such genetic differentiation exist, relative contributions of alfalfa- and back locust- associated populations could be inferred with potential implications for virus transmission. I used single nucleotide polymorphisms (SNPs) to compare genetic structure across populations of cowpea aphids on alfalfa, on black locust, and trapped individuals flying into focal pumpkin fields. Alfalfa and black locust populations were sampled at local and regional scales to examine the relationship of geographic distance with genetic structure, and host-associated populations compared to identify putative host-associated genotypes. I predicted that population genotypes would positively correlate within local-scale geographic distance rather than regional- scale, and that individuals collected from alfalfa and black locust trees would represent distinct genotypes. 3.2 Methods Aphid Collection I trapped alate aphids landing in four focal pumpkin fields with pans (DiFonzo et al. 1997) in a 1:4 propylene glycol and water solution, collecting aphids once a week. Traps were set out for 14 weeks after pumpkin plant emergence or transplant in fields beginning the week of June 24, 2013, and ending with the onset of plant senescence in the week of September 23, 2013. Focal fields were located in Northwest Indiana in Tippecanoe, White, and Jasper Counties (see GPS coordinates, Table 3-1). I identified cowpea aphids (Aphis craccivora Koch) from all alates trapped weekly within each field, and stored them separately by field and collection date at -80° C in undiluted ethanol. I collected apterae by hand from potential source populations on black locust (Robinia pseudoacacia L.) and with sweep nets in alfalfa (Medicago sativa L.). Collections from black locust or alfalfa populations occurred within the duration of focal field sampling. Any

temporal variation between alfalfa or black locust population collection dates largely result from variable periods of aphid colonization, as collection attempts were made throughout the sampling period until at least 20 individuals were collected in a single sampling event. Source population locations were selected to represent a wide variety of distances both locally (<50 km) and regionally (>100 – 500 km) relative to the four focal pumpkin fields. To include a more representative composition of aphid clones, or multilocus genotypes (MLGs), per population, I swept alfalfa fields 10 times in 10 locations at least 30 m apart. Populations on locust were often concentrated on new seedling growth. Because populated seedlings were isolated, cowpea aphids in locust populations were collected from a single source. All aphid samples were stored at -80°C in undiluted ethanol. Individual aphid DNA was extracted with DNeasy Blood and Tissue Kit (Qiagen, Hilden, Germany), and samples with DNA concentration <5 ng μl-1 amplified with REPLI-g Mini Kit (Qiagen, Hilden, Germany). In total, I extracted DNA from 465 aphids.

Table 3-1. Collection location, date, population type, and sample size of cowpea aphids used in 2013 analyses. Population Population Type Date N Latitude Longitude K migrant 24-Jun–27-Sep 11 40.605161 -86.900153 M migrant 24-Jun–27-Sep 15 40.295096 -86.903664 O migrant 24-Jun–27-Sep 5 40.520730 -86.818576 Mt migrant 24-Jun–27-Sep 9 41.015903 -87.218693 IL alfalfa 13-Aug 19 40.908390 -90.231950 WI alfalfa 12-Aug 20 42.698720 -89.826590 MI alfalfa 25-Aug 19 42.169160 -85.941100 sIN alfalfa 11-Sep 14 38.781800 -86.621310 wcIN alfalfa 11-Sep 20 39.629390 -85.314754 nearMT alfalfa 20-Aug 20 41.099210 -87.075853 nearM alfalfa 16-Jul 20 40.338240 -86.646870 nearOK10 alfalfa 4-Sep 19 40.541460 -86.989624 nearOK8 alfalfa 20-Aug 18 40.589690 -86.964500 btwnMtK alfalfa 16-Sep 19 40.716740 -86.865237 MIbl black locust 23-Jun 18 41.784260 -85.343630 OH black locust 9-Sep 17 40.436230 -82.916479 KY black locust 28-Sep 20 37.993920 -84.526436 wcINbl black locust 11-Sep 17 39.764160 -85.417947 nearMTbl3 black locust 12-Jul 20 40.866810 -87.098423 CarolCo black locust 6-Sep 19 40.514070 -86.530552 nearOK black locust 15-Jul 20 40.502940 -86.868650 btwnMO black locust 4-Sep 16 40.426560 -86.902034

SNP Identification and Genotyping High-throughput sequencing (RAD-seq) of DNA from ten cowpea aphids generated restriction site associated DNA markers (RAD tags) (Molecular and Cellular Imaging Center, The Ohio State University, Wooster, OH). The RAD tags were then used to identify de novo candidate single nucleotide polymorphisms (SNPs) with the Stacks program (Catchen et al. 2011, 2013). Fifty candidate SNPs were run across aphid DNA samples, of which seven candidates failed validation (LGC Genomics LLC, Beverly, MA). Both aphid samples and loci with >10% missing data were identified with GenAlEx v6.5 program (Peakall & Smouse 2006, 2012) and manually discarded. Loci with a minor allele frequency <0.01 in one or more populations were discarded using the poppr package (Kamvar et al. 2014) and R v3.2.2 program (R Core Team 2015). Four additionally excluded loci were identified as nonneutral under balancing selection using LOSITAN program (Beaumont & Nichols 1996) (Fig. C3- 1). Ultimately, the dataset included 375 aphid individuals and 22 loci. The 22 loci used in analyses will be deposited in GenBank. Polymorphism, Genetic Diversity, and Group Assignment I used the Paetkau assignment test (Paetkau et al. 2004) in GenAlEx to assign samples to populations. GenAlEx also matched individual aphid samples to multi locus genotypes (MLGs); missing data can inflate the number of MLGs by reducing matches. To correct for this, I also manually assigned samples to MLG by ‘best estimate’, assigning an MLG only after eliminating all other possible MLG matches (Gitzendanner et al. 2012). To assess genetic diversity, I followed recommendations made by Arnaud-Haond et al. (2007) in selecting diversity parameters appropriate for clonal organisms: I used the poppr package to calculate Simpson complement (D*) and Simpson evenness indices (V), and the slope of the Pareto distribution (c), and calculated genotypic richness (R: R=(G-1)/(N-1), where G=number of MLGs and N=population size). Again following recommendations for analysis of clonal organisms by Arnaud-Haond et al. (2007), I recalculated all of the statistical analyses that follow after removing all repeat MLGs within a population to exclude the effect of clonality unless otherwise stated. Observed (Ho) and expected heterozygosity (He), inbreeding coefficient (Fis), allele frequencies, and deviation from Hardy- Weinberg Equilibrium were calculated with GenAlEx. Also in GenAlEx, I ran a Mann-Whitney U test to compare population parameters among focal-field, alfalfa- and black locust-associated populations. Although varying sample collection dates can lead to variations in self-population assignment, with clonal diversity within populations either increasing (Michel et al. 2009,

Orantes et al. 2012) or decreasing (Komazaki et al. 2011, Sunnucks et al. 1997) over time, levels of self-population assignment in this study were very low across all populations (Table 3-2). As such, I chose not to assess frequency of self-assignment as a function of collection date. However, I assessed the temporal relationship of host-associated MLGs in focal field alatae with a Chi- square Test for Independence on the number of early and late alatae occurring in all four focal fields as grouped by alfalfa-association, black locust-association, or lack of association. The early time bin contained samples collected weeks 1–7 of the sampling period, and the late bin, weeks 8–14. All individuals in focal fields were included in the chi-square analysis. Spatial Patterns and Population Genetic Structure The relationship of geographical distance between sample sites was assessed with spatial autocorrelation in GenAlEx, generating 999 bootstrapped coefficient values (r) with 999 random permutations. Geographic distances of 10, 15, 20, 25, 50, 100, 200, and 400 km were included in analyses. Because I expected local populations to be more genetically similar, incremental distance units were much closer together in identify the most influential scale within 50 km. Beyond the local scale I expected correlation to be uniformly weak, and thus included only three exponentially increasing distances that spanned the remainder of population sites. A principal coordinates analysis (PCoA) generated population clusters based on genetic similarity using the Nei distance matrix in GenAlEx. To further visualize genetic structure among and within aphid populations, I used STRUCTURE v2.3.4, which infers populations among samples and can identify alatae or admixed samples (Pritchard et al. 2000, Falush et al. 2003). I used a burn-in length of 50,000 with an admixed ancestry model and 100,000 Monte Carlo Markov-Chain repetitions to test K=2–22 number of genetic groups. Each value of K was replicated ten times. The probable value of K was generated using the Evanno et al. (2005) method with STRUCTURE HARVESTER program (Earl 2012). The graphical output was visualized with Distruct v1.1 (Rosenberg 2004). Because STRUCTURE methods assume HI, only the dataset with all repeat MLGs within populations underwent the analysis. 3.3 Results Polymorphism and Genetic Diversity Before the removal of repeat MLGs, all loci in 11 populations were not in HI, including all alfalfa-associated populations and one locust-associated population (Table C3-1). The remaining locust populations had almost entirely monomorphic loci. After best estimate repeat MLGs were removed there was only a single locus in a focal field population, K, not in equilibrium or monomorphic (Table C3-2). Six repeat MLGs were identified, with memberships

of 201, 134, 2, 3, 4, and 8 individuals. The remaining 23 MLGs occurred in only one individual across all populations. The largest MLG was primarily associated with alfalfa populations, except several black locust samples collected at a site in Ohio. The second largest MLG was primarily associated with black locust populations, except one individual from an alfalfa site from West- Central Indiana. Of the remaining repeat MLGs, one was associated with black locust, one with alfalfa, and two were found only in alate aphids trapped in focal fields. Overall, levels of genotypic diversity were low, with focal field alate populations and population OH exhibiting the highest relative diversity (Table 3-2).

Table 3-2.Individual sample assignments to self (self P) or other populations (other P), number of matching, unique, and total MLGs within each population, and population genetic diversity parameters genotypic richness (R), Simpson complement (D1), Simpson evenness (V), and the slope of the Pareto distribution (c).

ASSIGNMENT GENOTYPES Pop. self P other P self P/N # match # unique total R D1 V c K 0 11 0 3 5 8 0.70 0.14 -10.48 2.00 M 7 8 0.47 4 4 8 0.50 0.16 -2.79 1.46 O 0 5 0 1 3 2 0.25 0.28 -0.53 2.00 Mt 0 9 0 3 4 7 0.75 0.16 -15.88 1.81 IL 19 0 1 1 0 1 0 0.72 —a 0.50 WI 0 20 0 1 1 2 0.05 0.73 1.48 0.49 MI 0 19 0 1 0 1 0 0.72 —a 0.50 sIN 0 14 0 1 0 1 0 0.74 —a 0.44 wcIN 0 20 0 2 0 2 0.05 0.44 0.78 0.81 nearMT 0 20 0 1 0 1 0 0.74 —a 0.37 nearM 0 20 0 2 1 3 0.11 0.65 0.90 0.58 nearOK10 0 19 0 1 0 1 0 0.90 —a 0.24 nearOK8 0 18 0 2 0 2 0.06 0.64 1.25 0.42 —a 0.24 btwnMtK 0 19 0 1 0 1 0 0.90 —a 0.24 MIbl 18 0 1 1 0 1 0 0.90 OH 0 17 0 1 4 5 0.25 0.37 -0.13 0.90 KY 0 20 0 2 0 2 0.05 0.58 1.11 0.66 wcINbl 0 17 0 2 0 2 0.06 0.69 1.38 0.53 nearMT3 0 20 0 2 0 2 0.05 0.91 1.89 0.24 CarolCo 0 19 0 1 0 1 0 0.90 —a 0.24 nearOK 0 20 0 2 0 2 0.05 0.91 1.89 0.24 btwnMO 0 16 0 1 1 2 0.07 0.77 1.59 0.42 a=not applicable due to monomorphic population (total # MLGs=1)

Total number of MLGs and number of unique MLGs were highest in focal field populations and population OH, with remaining populations containing one or two matching MLGs and 0 or 1 unique. The total relative proportion of black locust-associated, alfalfa-associated, and unassociated MLGs within focal field populations shifted significantly between early- and late- season (χ2=7.35, df=2, n=40, P<0.05). Black locust-associated MLGs occurred primarily in early- season individuals and alfalfa-associated MLGs primarily in late-season (Table 3-3).

Table 3-3.Total number of alate cowpea aphids trapped in Lower Northwestern Indiana focal pumpkin fields, and percentage MLGs associated with alfalfa or black locust populations early and late in the growing season (early: weeks 1–7; late: weeks 8–14). EARLY LATE Focal % locust- % alfalfa- % locust- % alfalfa- Population Nearly associated associated Nlate associated associated K 7 57.14 0.00 4 0.00 0.25 M 12 33.33 0.00 3 0.00 66.67 O 5 0.00 0.40 0 0.00 0.00 Mt 2 100.00 0.00 7 0.14 28.57

Spatial Patterns and Population Genetic Structure Within the full dataset, strong significant positive spatial autocorrelation (r=0.75) occurred within 10 km (Fig. 3-1A). The relationship was reversed, with a significant negative correlation between genetic relatedness and geographic distance of moderate strength at 15 km (r=-0.34), decreasing in strength at 20 km (r=-0.15), 25 km (r=-0.26), and 50 km (r=-0.06), and nonsignificant at all larger distance classes. Within the dataset excluding the effect of clonality, correlation values were uniformly low and nonsignificant (Fig. 3-1B).

1.000 0.800 0.600

0.400

r 0.200 r 0.000 +95% -0.200 -0.400 -95% -0.600 10 15 20 25 50 100 200 400 A) Distance Class (End Point)

0.300 0.200

0.100

r 0.000 r -0.100 +95% -0.200 -95% -0.300 10 15 20 25 50 100 200 400 B) Distance Class (End Point)

Figure 3-1. Spatial autocorrelation analyses of genetic relatedness by geographic distance: A) full dataset, B) dataset excluding effects of clonality.

Variation among populations was largely captured by the first principal component (PC 1) in both PCoAs (Fig. 3-2A,B). Excluding repeat MLGs resulted in an increase in variation capture by PC 2, but the percentage variation explained was nevertheless relatively minor (12%). Both PCoAs displayed three major clusters, with all black locust populations excluding OH clustering at one end of the PC 1 axis, most alfalfa populations clustering at the opposite end of the PC 1 axis, and focal field populations clustering in the middle. Populations plotted at ca. 11- 12 independent positions along axes.

-0.080 -0.040 0.000 0.040 0.020

M K OH 0.010 O nearOK8 Mt

nearM

3% Variation 3%

– WI 0.000 MIbl wcIN btwnMO sIN PC 2 2 PC nearOKbl btwnMtK CarolCo nearOK10 nearMT3 nearMT wcINbl MI -0.010 KY IL PC 1 – 96% Variation A)

-0.080 -0.040 0.000 0.040 0.080 0.030

OH 0.020 M K O nearOK8 MT WI

0.010

12% Variation 12%

– nearM 0.000 PC 2 2 PC MIbl KY IL btwnMO MI wcINbl wcIN sIN -0.010 nearMT3 nearMT CarolCo PC 1 – 87% Variation nearOK10 B) nearOKbl btwnMtK

Figure 3-2. Principal coordinates analyses grouping populations by genetic similarity, based on Nei distance matrices: A) including all aphid samples, B) excluding the effect of clonality. = focal field alates, = black locust population, =aphid population

The STRUCTURE analysis of populations excluding repeat MLGs suggested a likelihood of K=10 distinct genetic groups (Fig. 3-3). The STRUCTURE diagrams for K=10 is also displayed below (Fig. 3-4A). The STRUCTURE diagram illustrates virtual monomorphism among black locust populations, with the exception of OH which most closely matches the genetic composition of alfalfa populations. The black locust-associated group (indicated in orange) is present in smaller proportions in two alfalfa populations, nearM and wcIN, as well as in focal field populations. Alfalfa populations are composed primarily of seven or eight groups, also found in focal field populations. Although manual assignment of alfalfa- or black locust- associated MLGs suggested approximately equal occurrence of host-associated MLGs within pan populations, the diagrams suggest a greater proportion of individuals trapped in focal fields fall within alfalfa-associated genetic groups than the black locust group. Results should be interpreted with caution; although the program ignores missing data, group estimates for samples with missing data are less accurate (Pritchard et al. 2010).

Figure 3-3. The relative likelihood of K=2–22 groups derived by the Evanno method, as indicated by ΔK.

Figure 3-4. STRUCTURE diagram of K=10 genetic groups by population (indicated below graph) and population type (indicated above graph: M=alate aphids trapped in focal fields, A=population collected from alfalfa, L=population collected from black locust).

3.4 Discussion Results support a pattern of genetic relatedness among populations within 10 km geographic distance, and reduced gene flow beyond 15 km. Because heterozygosity was so low among samples, excluding clonality by removing repeat MLGs within populations rendered

statistical power very low, making it difficult to assess spatial autocorrelation among populations. Results of the spatial analysis on the full dataset are plausible, however, and reflect the relative frequency of local over long-distance dispersal (Loxdale et al. 1993). This could be clarified by repeating the study with more populations; specifically, sampling additional focal fields and putative populations within a 10 km radius surrounding focal fields, given the positive correlation between genetic relatedness and this spatial scale with the full dataset. Examining spatial autocorrelation among populations at a finer scale within the 10 km range would also be important to assess the specific degree of local movement. Individual membership of genetic group clearly varied by host association, making identification of an alate’s likely host-plant possible. The temporal variation in proportion of alates trapped in the focal fields with alfalfa- versus black locust-associated MLGs suggests a pattern of cowpea aphid immigration from black locust populations early in the season and late in the season from alfalfa populations. If the pattern is consistent across years it has important implications for virus management. Crops are more susceptible to reductions in yield due to virus infection in earlier stages of growth; thus, vectors interacting with crops earlier in the growing season are of greater concern. Additionally, vector biotypes have been known to exhibit differential virus transmission efficiency (e.g., McGrath & Harrison 1995). The host-associated cowpea aphid MLGs could likewise differ in their ability to transmit plant viruses. The frequency of unique MLGs within focal field alates suggests they are also arriving from other populations, or even host-plants other than alfalfa or black locust. One possible alternative leguminous host- plant is soybean (Glycine max L.). Cowpea aphids have been observed colonizing soybeans (Blackman 2015), and it is a dominant crop in the Midwestern landscape. Other naturally- occurring populations of legumes such as clover (Trifolium spp.) and vetch (Vicia sativa L.) are possible alternative host-plants, as are many non-legumes, which cowpea aphids are known to colonize under more arid summer conditions (Blackman 2015). For example, cowpea aphids were identified colonizing apples (Malus domestica Borkh.) adjacent to focal field M (personal observation). Genetic grouping by black locust- or alfalfa-association is compelling in light of recent support for distinct locust- and alfalfa-biotypes (Wagner et al. 2015). Membership was not completely divided by alfalfa- or black locust-association, with a few individuals in one host population exhibiting a genotype associated with the other, as might be expected with frequent local movement among populations. Because the host-associated biotypes are mediated by endosymbiont association, it is possible that the MLG differentiation is coincidental and not

reflective of adaptation. Horizontal transfer of the endosymbiont would facilitate a sustained relationship between clonal populations and host type, however, and could result in the accumulation of adaptive mutations in an MLG. Overall, I observed a pattern of distinct genetic grouping by population association with alfalfa or black locust, and temporal patterns of host-associated cowpea aphid alate arrival in pumpkin fields. Additionally, populations likely genetically differentiate at >15 km, although the low genetic diversity within populations make interpretation of results difficult and require a larger sample size. Results pave the way for future investigations on MLG interactions with endosymbiont association on host performance and biotype formation, as well as on the relative importance of alfalfa or black locust as vectors based on seasonal abundance or vectorial capacity.

3.5 References

Arnaud-Haond, S., C. M. Duarte, F. Alberto, and E. A. Serrão. 2007. Standardizing methods to address clonality in population studies. Molecular Ecology 16: 5115-5139. Beaumont, M. A., and R. A. Nicols. 1996. Evaluating loci for use in the genetic analysis of population structure. Proceedings of the Royal Society B: Biological Sciences 263: 1619- 1626. Blackman, R. L. Aphids on the World's Plants. Natural History Museum, London. Blackman, R. L., and V. F. Eastop. 1984. Aphids on the world's crops. An identification and information guide, John Wiley. Blackman, R. L., and V. F. Eastop. 2008. Aphids on the world's herbaceous plants and shrubs, John Wiley & Sons. Catchen, J., P. A. Hohenlohe, S. Bassham, A. Amores, and W. A. Cresko. 2013. Stacks: an analysis tool set for population genomics. Molecular ecology 22: 3124-3140. Catchen, J. M., A. Amores, P. Hohenlohe, W. Cresko, and J. H. Postlethwait. 2011. Stacks: building and genotyping loci de novo from short-read sequences. G3: Genes, Genomes, Genetics 1: 171-182. Delmotte, F., N. Leterme, J. P. Gauthier, C. Rispe, and J. C. Simon. 2002. Genetic architecture of sexual and asexual populations of the aphid Rhopalosiphum padi based on allozyme and microsatellite markers. Molecular Ecology 11: 711-723. DiFonzo, C. D., D. W. Ragsdale, E. B. Radcliffe, N. C. Gudmestad, and G. A. Secor. 1997. Seasonal abundance of aphid vectors of potato virus Y in the Red River Valley of Minnesota and North Dakota. Journal of Economic Entomology 90: 824-831. Earl, D. A. 2012. STRUCTURE HARVESTER: a Ibsite and program for visualizing STRUCTURE output and implementing the Evanno method. Conservation Genetics Resources 4: 359-361. Evanno, G., S. Regnaut, and J. Goudet. 2005. Detecting the number of clusters of individuals using the software STRUCTURE: a simulation study. Molecular ecology 14: 2611-2620. Falush, D., M. Stephens, and J. K. Pritchard. 2003. Inference of population structure using multilocus genotype data: linked loci and correlated allele frequencies. Genetics 164: 1567-1587.

Fuller, S., P. Chavigny, L. Lapchin, and F. Vanlerberghe‐Masutti. 1999. Variation in clonal diversity in glasshouse infestations of the aphid, Aphis gossypii Glover in southern France. Molecular Ecology 8: 1867-1877. Gitzendanner, M. A., C. W. Iekley, C. C. Germain-Aubrey, D. E. Soltis, and P. S. Soltis. 2012. Microsatellite evidence for high clonality and limited genetic diversity in Ziziphus celata (Rhamnaceae), an endangered, self-incompatible shrub endemic to the Lake Wales Ridge, Florida, USA. Conservation Genetics 13: 223-234. Guillemaud, T., L. Mieuzet, and J. Simon. 2003. Spatial and temporal genetic variability in French populations of the peach–potato aphid, Myzus persicae. Heredity 91: 143-152. Kamvar, Z. N., J. F. Tabima, and N. J. Grünwald. 2014. Poppr: an R package for genetic analysis of populations with clonal, partially clonal, and/or sexual reproduction. PeerJ 2: e281. Komazaki, S., S. Toda, T. Shigehara, S. Kanazaki, H. Izawa, K. Nakada, and E. Souda. 2011. The genetic structure of Aphis gossypii populations in Japanese fruit orchards. Entomologia Experimentalis et Applicata 140: 171-179. Loxdale, H. D., J. Hardie, S. Halbert, R. Foottit, N. A. C. Kidd, and C. I. Carter. 1993. The relative importance of short- and long-range movement of flying aphids. Biological Reviews 68: 291-311. McGrath, P. F., and B. D. Harrison. 1995. Transmission of tomato leaf curl geminiviruses by Bemisia tabaci: effects of virus isolate and vector biotype. Annals of Applied Biology 126: 307-316. Michel, A. P., W. Zhang, J. K. Jung, S.-T. Kang, and M. A. R. Mian. 2009. Population genetic structure of Aphis glycines. Environmental Entomology 38: 11. Orantes, L., W. Zhang, M. Mian, and A. Michel. 2012. Maintaining genetic diversity and population panmixia through dispersal and not gene flow in a holocyclic heteroecious aphid species. Heredity 109: 127-134. Paetkau, D., R. Slade, M. Burden, and A. Estoup. 2004. Genetic assignment methods for the direct, real‐time estimation of migration rate: a simulation‐based exploration of accuracy and poIr. Molecular Ecology 13: 55-65. Paulsrud, B. Year. Published. Workshop Summary, p. 109. In, Midwest Pest Management Strategic Plan for Processing & Jack-o-Lantern Pumpkins: Illinois, Indiana, Iowa and Missmyi, 2005, Urbana, Illinois. University of Illinois Extension.

Peakall, R., and P. E. Smouse. 2006. Genalex 6: genetic analysis in Excel. Population genetic software for teaching and research. Molecular Ecology Notes 6: 288-295. Peakall, R., and P. E. Smouse. 2012. GenAlEx 6.5: genetic analysis in Excel. Population genetic software for teaching and research—an update. Bioinformatics 28: 2537-2539. Pritchard, J. K., M. Stephens, and P. Donnelly. 2000. Inference of population structure using multilocus genotype data. Genetics 155: 945-959. R Core Team. 2015. R : A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria. URL https://www.R-project.org/. Sandrock, C., J. Razmjou, and C. Vorburger. 2011. Climate effects on life cycle variation and population genetic architecture of the black bean aphid, Aphis fabae. Molecular Ecology 20: 4165-4181. Sunnucks, P., P. De Barro, G. Lushai, N. Maclean, and D. Hales. 1997. Genetic structure of an aphid studied using microsatellites: cyclic parthenogenesis, differentiated lineages and host specialization. Molecular Ecology 6: 1059-1073. Wagner, S. M., A. J. Martinez, Y. M. Ruan, K. L. Kim, P. A. Lenhart, A. C. Dehnel, K. M. Oliver, and J. A. White. 2015. Facultative endosymbionts mediate dietary breadth in a polyphagous herbivore. Functional Ecology 29: 1402-1410. Wilson, A., P. Sunnucks, R. Blackman, and D. Hales. 2002. Microsatellite variation in cyclically parthenogenetic populations of Myzus persicae in south-eastern Australia. Heredity 88: 258-266.

CHAPTER 4. FACULTATIVE ENDOSYMBIONTS ALTER FEEDING BEHAVIOR OF A STYLET-BORNE PLANT VIRUS VECTOR

4.1 Introduction Facultative endosymbionts, microbes housed internally in a sustained relationship with their host but not necessary for survival, are now implicated in mediating a wide range of physiological mechanisms from regulating immune response to altering mood in humans (e.g., Kelly et al. 2004, Forsythe et al. 2009). Less is known in the realm of plant-insect interactions, and most investigation thus far have focused on parasite defense, heat tolerance, and protection from pathogens (e.g., Oliver et al. 2005, Russell & Oliver 2006, Scarborough et al. 2005); research into facultative endosymbiont (hereafter symbiont) effects in plant pathogen epidemiology is scant, and none thus far involves stylet-borne, or nonpersistent viruses (e.g., Gottlieb et al. 2010, Rana et al. 2012). Recently, the first clear-cut case of herbivore host-plant range modification by symbiont association was documented (Wagner et al. 2015). It is unknown whether symbiont associations can also mediate herbivore feeding behavior, but the possibility could have substantial impact on the spread of plant pathogens such as nonpersistent viruses. The cowpea aphid, Aphis craccivora Koch, was recently discovered to exist in ‘host- associated biotypes’ driven by symbiont association (Wagner et al. 2015). In this system, aphid populations on black locust (Robinia pseudoacacia L.) are associated with Arsenophonus, and populations in alfalfa (Medicago sativa L.) with H. defensa (Brady et al. 2013; Brady & White 2013). Among aphids in the ‘locust-associated-biotype’, the symbiont Arsenophonus increases aphid performance on black locust while decreasing it on alternative host species. Within the ‘alfalfa-associated-biotype’, H. defensa association was found to decrease performance and survival on non-alfalfa plants (Wagner et al. 2012). This species of aphid has been implicated as an important vector of the nonpersistent virus, watermelon mosaic virus (WMV), in Midwestern pumpkins (Angelella et al. 2015), and it is likely that source populations of cowpea aphids occur on black locust and alfalfa, and that Arsenophonus or H. defensa symbiont associations occur in the cowpea aphids interacting with the pumpkin-nonpersistent virus system.

With such dietary restrictions resulting from symbiont association, concurrent symbiont- mediated modification in plant selection behavior would be adaptive for both symbiont and host, with ramifications for nonpersistent virus transmission. Aphid feeding behavior is characterized by a consistent order of probing behaviors in plant tissues with their mouthparts, with certain behaviors linked to nonpersistent virus acquisition and inoculation. Stylet routes begin in the epidermis and make contact with many cells in the mesophyll along the way to the vascular bundle and sieve elements, where phloem may eventually be ingested, but punctured cells seem to be kept alive by having cell wall breaches filled in with salivary sheath material (Tjallingii & Hogen Esch 1993). Because nonpersistent viruses are inoculated into host plant cells, probing in the epidermis and mesophyll are more likely to result in transmission or acquisition of virus, whereas sustained feeding from phloem sieve tubes is less likely to do so. Upon puncturing plant tissue, aphids salivate a proteinaceous, gel-like saliva, which surrounds the stylet like a sheath (Bennett 1934, Miles 1999). Aphids intermittently elute a more watery saliva as probing continues, pausing to ingest a bit of saliva back in as they move through the tissues, which likely facilitates contact with gustatory cues indicating host plant identity and quality (Wensler & Filshie 1969; Tjallingii 1995, reviewed in Miles 1999). The exuding and re-ingesting of saliva in plant cells is thought to facilitate nonpersistent virus transmission and acquisition (Martin et al. 1997). If feeding behavior changes by endosymbiont infection on a plant, virus acquisition will likely be facilitated by a lengthened period of shallow intracellular probes with the aphid stylet – behavior associated with host plant exploration and subsequent rejection (Powell 1991, Powell et al. 1995, Fereres & Moreno 2009). Few accounts of endosymbiont-mediated changes in arthropod behavior currently exist. One such example documented a decrease in ballooning, an aerial dispersal behavior, performed by the money spider, Erigone atra, when infected with the endosymbiont, Rickettsiella (Goodacre et al. 2006). In another example, the parasitoid wasp, Encarsia pergandiella, oviposited significantly fewer eggs into their primary insect host, when infected with Wolbachia (Zchori- Fein et al. 2001). These examples demonstrate that endosymbiont-induced behavioral changes occur, and are likely more common than currently appreciated. Using an aphid-virus-endosymbiont system, I explored whether interactions occurred to affect epidemiology via endosymbiont-associated mediation of feeding behavior (Fig. 4-1). I hypothesize that patterns of feeding behavior will vary with symbiont association, independent of host-plant origin or virus infection. One alternative possibility is that feeding behavior will vary by virus infection, but not symbiont association or host-plant origin. Research has documented

nonpersistent virus-mediated changes in plant physiology leading to changes in vector feeding behavior which enhance transmission, such as more frequent shallow probing and decreased phloem-feeding (reviewed in Mauck et al. 2012). It is plausible that WMV infection could similarly affect vector-plant interactions. A third possibility is that host-plant origin, and not endosymbiont association, will relate to variations in feeding behavior. The locust-origin biotype is driven by Arsenophonus-mediated performance changes on plant-host species. For the third possibility to occur, selection for behavioral changes in locust- or alfalfa-origin biotypes would have occurred independent of symbiont association. This is possible if symbiont associations are maintained across successive generations by high rates of maternal or horizontal transmission, which is more likely with Arsenophonus as H. defensa associations are more frequently lost from colonies (Dykstra et al. 2014, Wagner et al. 2015).

Feeding Endosymbiont-Insect Behavior Infection Association Status

Host Plant

Vector Competence

Figure 4-1. Graphical abstract of direct and indirect effects among host-plant association, vector, endosymbiont, and plant. 4.2 Methods Aphid Colonies Colonies were obtained from the White lab at the University of Kentucky, and included one locust-origin clone (LE) and one alfalfa-origin clone (AC) (see Wagner et al. 2015). Two isolines of each clone were maintained: LE infected with Arsenophonus (Ars+), LE cured (Ars-), AC infected with Hamiltonella defensa (Ham+), and AC cured (Ham-). The selective curing procedure was modified from Douglas et al. (2006) (see Wagner et al. 2015 for details), and symbiont associations were verified with PCR (Brady & White 2013). To mitigate the possibility of natal plant preference all isolines were maintained on the same universally-accepted nonhost, fava bean (Vicia fava L.). Isolines were propagated in 50 x 50 cm2 plexiglass cages with bridal

veil fabric screening (JoAnn Fabric, Lafayette, IN) on front and back for ventilation. Aphids were age-synchronized by isolating 4th instar larvae on fava bean leaf discs, and using only apterous adults found two days later. Virus Inoculations Watermelon mosaic virus was propagated in pumpkin (Cucurbita pepo L.) after mechanical inoculation from freeze-dried plant tissue (ATCC, Manassas, VA). Pumpkin plants were mechanically inoculated in the two-leaf cotyledon stage using inoculated pumpkin leaf tissue following protocol described by Eigenbrode et al. (2002). Sham inoculations using only carborundum powder and 0.1 M phosphate buffer solution of PH 7 were a control treatment. Virus inoculations were verified with a diagnostic WMV ELISA (Agdia® Inc., Elkhart, IN). Plants were used 2–3 weeks post-inoculation. Pumpkins were maintained in an insect-free growth chamber under conditions of 16 L:8 D, 25% r.h. and 23°C. Experimental plants were “Mystic plus” cultivar pumpkins with powdery mildew resistance (Harris® Seeds, Rochester, NY), propagated in autoclaved soil (Hummert International, Earth City, MO) and 6” pots (Hummert International, Earth City, MO) with 1 tsp. Osmocote® (The Scotts Company, Maryville, OH). Electrical Penetration Graphs Laboratory studies utilizing electrical penetration graph (EPG) systems allow for the quantification and comparison of nuanced feeding behaviors as they relate to virus transmission and acquisition. Aphids were affixed to an electrode attached to a thin gold wire with a droplet of silver glue (EPG Systems, the Netherlands), placed on the upper surface of the youngest fully developed leaf, and the other electrode inserted into the soil. The procedure was repeated until each of eight EPG probes was attached to an aphid. Plants and aphids were then placed into a Faraday cage. As the aphid probed and/or fed on the plant, a computer hooked up to the EPG system recorded wavelengths of electricity flowing through as the aphid-plant circuit closed. In initial exploratory EPG runs, aphids were maintained on plants for 8 h, providing ample time for sustained feeding to occur on a host plant (Schwarzkopf et al. 2013). However, no phloem feeding or sieve element location was observed, supporting an observed lack of pumpkin colonization (i.e., host acceptance followed by feeding and/or reproduction) by cowpea aphids (Angelella et al. 2015). As noncolonizers, they would therefore not be expected to remain on a pumpkin plant for an extended duration, and would rather pause to probe, reject the plant, and move on to find a host. Subsequent EPG analyses were run for the duration of 1 h. As no phloem- feeding events occurred, only parameters quantifying intracellular probes were measured as time to the first potential drop and frequency of potential drops within the first 15 min. (Fig. 4-2).

Measurements began from the moment aphids made contact with the leaf. This was replicated 15 times on each virus treatment for all four colonies (n= 15 plants x 4 colonies x 2 WMV- inoculated/control categories), using new aphids.

p a t h w a y phase B 50 sec A B C C F

pd Adapted from: W.F. Tjallingii, 2014 Figure. 4-2. Graphical output of electrical penetration graph (EPG) measurements. Potential drop (pd) indicates intraceullar puncture, behavior associated with nonpersistent virus acquisition and inoculation. Parameters measured by EPG assays included time to first pd, and number of pds occurring in the first 15 minutes.

Data Analysis Individual trials resulting in aphids disconnecting from the EPG apparatus, losing contact with the leaf, or aphid mortality were excluded from analyses. Resultant group sizes were as follows: total control n=25, total WMV-inoculated n=34, alfalfa-origin n=37, locust-origin n=38, locust-Ars n=19, locust-cured n=19; alfalfa-Ham n=17, alfalfa-cured n=20; locust-Ars/control n=9, locust-cured/control n=12, locust-Ars/WMV n=10, locust-cured/WMV n=7; alfalfa- Ham/control n=9, alfalfa-cured/control n=11, alfalfa-Ham/WMV n=8, alfalfa-cured/WMV n=9. Prior to analyses, time to probe initiation (first.pd) was square root transformed, and checked with Shapiro-Wilk’s test for normal distribution. Homogeneity of the variances were checked with Levene’s test, on first.pd by symbiont (Ars+ vs. Ars-, and Ham+ vs. Ham-) crossed with virus treatment (WMV-inoculated vs. control), or host-plant origin (locust vs. alfalfa) crossed with virus treatment. A three-way ANOVA with symbiont nested within host-plant origin was then conducted on transformed first.pd data, examining the differential effects of host-plant origin and symbiont association under virus treatment conditions. The number of potential drops in the first 15 min. (pd.15) data were highly skewed, and thus analyzed with Poisson regression to examine a three-way interaction of effects among host-plant origin, symbiont and virus treatment, with symbiont nested within host-plant origin. 4.3 Results & Discussion Speed of probe initiation (first.pd) did not differ by host-plant origin, symbiont association, or virus treatment (Table 4-1). However, several treatments were able to predict

probing frequency variation (Table 4-2): aphids probed more frequently on virus-infected plants (Fig. 4-3), and there was a host-plant origin by symbiont interaction whereby locust-origin aphids infected with Arsenophonus probed less frequently than those cured of symbiont association, and alfalfa-origin aphids infected with H. defensa probed more frequently than cured aphids (Fig. 4- 4A,B).

Table 4-1. Three-way ANOVA with symbiont (symb) nested within host-plant origin (origin) on square- root transformed time to probe initiation, examining effects of host-plant origin and symbiont association under virus treatment (virus) conditions (WMV-inoculated or mock-inoculated). Df SS Mean SS F P Origin 1 4 4.38 0.07 >0.05 Virus 1 1 0.66 0.01 >0.05 Origin x Virus 1 3 3.08 0.05 >0.05 (Origin:Symb) 2 64 31.76 0.50 >0.05 (Origin:Symb) x Virus 2 4 1.96 0.03 >0.05 Residuals 53 3362 63.43

Table 4-2. Three-way Poisson regression with symbiont (symb) nested within host-plant origin (origin) on number of probes in the first 15 min. (pds.15), examining effects of host-plant origin and symbiont association under virus treatment (virus) conditions (WMV-inoculated or mock-inoculated). Estimate SE z P *intercept 1.00 0.18 5.50 <0.001 Origin 0.41 0.25 1.67 0.09 *virus 0.49 0.24 2.02 0.04 origin x virus -0.42 0.33 -1.28 0.20 (locust-origin:symb) 0.39 0.20 1.94 0.05 *(alfalfa-origin:symb) 0.51 0.24 2.14 0.03 (locust-origin:symb) x virus 0.48 0.28 0.75 0.08 (alfalfa-origin:symb) x virus 0.09 0.31 0.29 0.77 RSD 275.10 Df 67

Figure 4-3. Overall effect of virus treatment (control, WMV-infected) on cowpea aphid feeding behavior on pumpkins.

Figure 4-4. Interaction effect between symbiont association and virus treatment on frequency of intracellular probes, by host-plant origin: A) locust-origin cowpea aphids, B) alfalfa-origin cowpea aphids.

These results suggest symbiont association can modify herbivore feeding behavior, and that differential modifications in probing frequency resulting from associations would in turn differentially affect likelihood of nonpersistent virus transmission. Greater frequency of intracellular probes in alfalfa-origin cowpea aphids with H. defensa association would increase likelihood of acquisition or inoculation, while the contrasting decreased frequency in locust-

origin aphids would render them less effective vectors. Furthermore, although aphids overall increased probing frequency on virus-infected pumpkins, there was some suggestion of a varied response to virus infection by symbiont species. Interactive symbiont-by-virus treatment effects on probing frequency approached significance in locust-origin aphids, but not in alfalfa-origin aphids (Figs. 4A,B). Virus-mediated changes in vector feeding behavior have been well documented, often leading to an enhanced propensity for transmission to occur (e.g., Ingwell et al. 2012, Mauck et al. 2012). If, however, aphids containing Arsenophonus do not alter feeding behavior on infected plants that could indicate decreased potential for virus acquisition and importance in cucurbit virus’ epidemiology. Although a mechanism has not yet been described for symbiont-mediation of host plant preference or performance in aphids, enterobacterial symbionts found in oral secretions of Colorado potato beetle larvae (Leptinotarsa decemlineata) were shown to interfere with herbivore-induced defenses, maintaining plant palatability for their feeding larval hosts (Chung et al. 2013). Were symbiont-plant defense interactions similar to those recently discovered occurring via L. decemlineata larval oral secretions to occur with Arsenophonus or H. defensa, it is plausible that subsequent changes in plant chemistry could affect aphid feeding behavior and vector competence. Compellingly, Arsenophonus has been isolated from the salivary glands of a close aphid relative (Rana et al. 2012); it is possible aphids could as well and thus exude bacteria in their saliva. Furthermore, as they probe plant tissue, aphids intermittently salivate, pausing to ingest a bit of saliva back in as they move through the tissues, likely facilitating contact with gustatory cues indicating plant identity and quality (Wensler & Filshie 1969, Tjallingii 1995, reviewed in Miles 1999). Aphid saliva could thus provide a conduit for plant-endosymbiont interactions, similar to those found in by Chung et al. (2013) in L. decemlineata larvae. However, the timescale for plant-symbiont interactions to induce behavioral changes in this study was <15 min., raising the question of whether interactions could occur quickly enough to be a mechanism driving results. Thus far, the influence of such symbionts on plant virus spread has only been investigated with tomato yellow leaf curl virus transmission by whiteflies (Bemisia tabaci) associated with H. defensa (Gottlieb et al. 2010). Additionally, molecular analyses showed an interaction between cotton leaf curl virus and a protein derived from Arsenophonus housed in the salivary glands of the most efficient vector biotype of whiteflies (Rana et al. 2012). The latter two studies imply vector competence may be enhanced in individuals containing the symbionts, but it is risky to generalize results as the study systems are very different. Whiteflies are close relatives

of aphids belonging to the same suborder (Hemiptera: Sternorrhynca), but the viruses were persistently transmitted and circulated within the insect, plausibly facilitating interaction with symbiont-derived virus-protective proteins; nonpersistent viruses adhere only temporarily to the distal tip of the stylet. Conclusions Feeding behavior in the cowpea aphid was differentially modified by symbiont association. Cowpea aphids with the locust-origin genotype probed less frequently when they contained the symbiont Arsenophonus, while aphids with the alfalfa-origin genotype probed more frequently when containing H. defensa. This could indicate changes in the likelihood of nonpersistent virus transmission by vectors depending on symbiont association: higher rates of probing would result in greater likelihood of virus acquisition and inoculation. Overall, there was a significantly greater frequency of intracellular probes on WMV-infected pumpkins, reinforcing the body of work reporting transmission-enhancing nonpersistent virus manipulation of vector behavior. A trend toward significance suggesting no response to the virus treatment among Arsenophonus-associated aphids should be further investigated, as that would additionally impact vector propensity. Additionally, assays should be conducted using additional genotypes associated with Arsenophonus and H. defensa to verify results, and transmission efficiency assays should be performed to validate the connection between symbiont-associated feeding behavior differentiation and virus infection dynamics.

4.4 References

Angelella, G.M., D.S. Egel, J.D. Holland, J.A. Nemacheck, C.E. Williams, and I. Kaplan. 2015. Differential life history trait associations of aphids with nonpersistent viruses in cucurbits. Environmental Entomology 44: 562-573. Bennett, C. W. 1934. Plant tissue-relations of the sugar-beet curly-top virus. Jmynal of Agricultural Research 48: 665-701. Brady, C., M. Asplen, N. Desneux, G. Heimpel, K. Hopper, C. Linnen, K. Oliver, J. Wulff, and J. White. 2013. Worldwide populations of the aphid Aphis craccivora are infected with diverse facultative bacterial symbionts. Microbial Ecology 67: 1-10. Brady, C. M., and J. A. White. 2013. Cowpea aphid (Aphis craccivora) associated with different host plants has different facultative endosymbionts. Ecological Entomology 38: 433-437. Chung, S. H., C. Rosa, E. D. Scully, M. Peiffer, J. F. Tooker, K. Hoover, D. S. Luthe, and G. W. Felton. 2013. Herbivore exploits orally secreted bacteria to suppress plant defenses. Proceedings of the National Academy of Sciences 110: 15728-15733. Drès, M., and J. Mallet. 2002. Host races in plant–feeding insects and their importance in sympatric speciation. Philosophical Transactions of the Royal Society of London. Series B: Biological Sciences 357: 471-492. Eigenbrode, S. D., H. Ding, P. Shiel, and P. H. Berger. 2002. Volatiles from potato plants infected with potato leafroll virus attract and arrest the virus vector, Myzus persicae (Homoptera: Aphididae). Proceedings of the Royal Society of London. Series B: Biological Sciences 269: 455-460. Fereres, A., and A. Moreno. 2009. Behavioural aspects influencing plant virus transmission by homopteran insects. Virus Research 141: 158-168. Forsythe, P., N. Sudo, T. Dinan, V. H. Taylor, and J. Bienenstock. 2010. Mood and gut feelings. Brain, Behavior, and Immunity 24: 9-16. Goodacre, S. L., O. Y. Martin, C. G. Thomas, and G. M. Hewitt. 2006. Wolbachia and other endosymbiont infections in spiders. Molecular Ecology 15: 517-527.

Gottlieb, Y., E. Zchori-Fein, N. Mozes-Daube, S. Kontsedalov, M. Skaljac, M. Brumin, I. Sobol, H. Czosnek, F. Vavre, and F. Fleury. 2010. The transmission efficiency of

tomato yellow leaf curl virus by the whitefly Bemisia tabaci is correlated with the presence of a specific symbiotic bacterium species. Journal of Virology 84: 9310-9317. Ingwell, L. L., S. D. Eigenbrode, and N. A. Bosque-Perez. 2012. Plant viruses alter insect behavior to enhance their spread. Scientific Reports 578: 1-6. Kelly, D., J. I. Campbell, T. P. King, G. Grant, E. A. Jansson, A. G. Coutts, S. Pettersson, and S. Conway. 2004. Commensal anaerobic gut bacteria attenuate inflammation by regulating nuclear-cytoplasmic shuttling of PPAR-γ and RelA. Nature Immunology 5: 104-112. Martin, B., J. L. Collar, W. F. Tjallingii, and A. Fereres. 1997. Intracellular ingestion and salivation by aphids may cause the acquisition and inoculation of non-persistently transmitted plant viruses. Journal of General Virology 78: 2701-2705. Mauck, K. E., N. A. Bosque-Perez, S. D. Eigenbrode, C. M. De Moraes, and M. C. Mescher. 2012. Transmission mechanisms shape pathogen effects on host-vector interactions: evidence from plant viruses. Functional Ecology 26: 1162-1175. Miles, P. W. 1999. Aphid saliva. Biological Reviews of the Cambridge Philosophical Society 74: 41-85. Oliver, K. M., N. A. Moran, and M. S. Hunter. 2005. Variation in resistance to parasitism in aphids is due to symbionts not host genotype. Proceedings of the National Academy of Sciences of the United States of America 102: 12795-12800. Powell, G. 1991. Cell membrane punctures during epidermal penetrations by aphids: consequences for the transmission of two potyviruses. Annals of Applied Biology 119: 313-321. Powell, G., T. Pirone, and J. Hardie. 1995. Aphid stylet activities during potyvirus acquisition from plants and an in vitro system that correlate with subsequent transmission. European Journal of Plant Pathology 101: 411-420. Rana, V. S., S. T. Singh, N. G. Priya, J. Kumar, and R. Rajagopal. 2012. Arsenophonus GroEL Interacts with CLCuV and Is Localized in Midgut and Salivary Gland of Whitefly B. tabaci. PloS one 7: e42168. Russell, J. A., and N. A. Moran. 2006. Costs and benefits of symbiont infection in aphids: variation among symbionts and across temperatures. Proceedings of the Royal Society B: Biological Sciences 273: 603-610. Scarborough, C. L., J. Ferrari, and H. Godfray. 2005. Aphid protected from pathogen by endosymbiont. Science 310: 1781-1781.

Schwarzkopf, A., D. Rosenberger, M. Niebergall, J. Gershenzon, and G. Kunert. 2013. To feed or not to feed: plant factors located in the epidermis, mesophyll, and sieve elements influence pea aphid’s ability to feed on legume species. PloS One 8: e75298. Tjallingii, W. 1995. Regulation of phloem sap feeding by aphids, pp. 190-209, Regulatory Mechanisms in Insect Feeding. Springer. Tjallingii, W. F., and T. Hogen Esch. 1993. Fine structure of aphid stylet routes in plant tissues in correlation with EPG signals. Physiological Entomology 18: 317-328. Wagner, S. M., J. A. White, and J. McCord. Year. Published. Endosymbiont effects on host plant usage by Aphis craccivora. In, Entomological Society of America 60th Annual Meeting, 2012, Knoxville, TN. Wagner, S. M., A. J. Martinez, Y. M. Ruan, K. L. Kim, P. A. Lenhart, A. C. Dehnel, K. M. Oliver, and J. A. White. 2015. Facultative endosymbionts mediate dietary breadth in a polyphagous herbivore. Functional Ecology 29: 1402-1410. Wensler, R. J. D., and B. K. Filshie. 1969. Gustatory sense organs in the food canal of aphids. Journal of Morphology 129: 473-492. Zchori-Fein, E., Y. Gottlieb, S. Kelly, J. Brown, J. Wilson, T. Karr, and M. Hunter. 2001. A newly discovered bacterium associated with parthenogenesis and a change in host selection behavior in parasitoid wasps. Proceedings of the National Academy of Sciences 98: 12555-12560.

APPENDICES

Appendix A: Chapter 1 Supplemental Material

Table A1-1. Primers used in the multiplex-RT-PCR assay to detect the targeted viruses. NCBI Primer 5’ Tm Amp. Position (bp) Virus Accession Sequence (°C)

F CGAAAGCTGTGAAGGGAACC 6988 59.1 ZYMV JN192422.1 1069 R GCCGCTATCCTCATCTTTGACTG 8056 61.4

F CATGGAAGCTAAGGTGATGGA 1982 57.4 CMV NC_002035 887 R AGCTGGATGGACAACCCGT 2868 61.2

F GCTAGTGACGGAAACGATGTGTC 9362 61.8 PRSV DQ374152.1 599 R CGAGCCCTATCAGGTGTTTT 9960 57.6

F GTATGGGTCCACCGCAGTAAAG 9767 61.0 WMV NC_006262.1 258 R CATTACCGTACCTCGGCTATT 10024 57.3 bp, measurement of amplicon length (amp.) in number of nucleotide base pairs

Table A1-2. Cumulative aphids caught in pan traps (n=5 traps/field) within fields monitored (2010: n=13, 2011: n=16) throughout the growing season (7 wks, Jul–Sep).

Subfamily Tribe Species 2010 2011

Anoeciinae — Anoecia cornicula (Walsh) 3 0

Anoecia eonothera Wilson 1 0

Anoecia setariae Gillette & Palmer 4 0

Anoecia spp. 4 10

Aphidinae Aphidini Aphis carduella Walsh 0 1

Aphis craccivora Koch 34 257

Aphis fabae Scopoli 5 7

Aphis glycines Matsumura 34 15

Aphis gossypii Glover 2234 359

Aphis lacinariae Gillette & Palmer 3 3

Aphis maculatae Oestlund 2 0

Aphis nasturtii Kaltenbach 1 0

Aphis nerii Bover de Fonscolombe 9 19

Aphis sedi Kaltenbach 2 0

Aphis spiraecola Patch 2 0

Aphis spp. 3 3

Aphis vernoniae Thomas 0 1

Hyalopterus pruni (Geoffroy) 4 0

Hysteroneura setariae (Thomas) 1 0

Protaphis middletonii (Thomas) 5 3

Rhopalosiphum maidis (Fitch) 35 20

Rhopalosiphum padi (Linnaeus) 416 29

Rhopalosiphum rufiabdominalis (Sasaki) 2 4

Schizaphis graminum (Rondani) 7 0

Macrosiphini Acyrthosiphon pisum (Harris) 4 9

Acyrthosiphon spp. 0 2

Brachycaudus spp. 0 1

Bracycorynella spp. 0 1

Capitophorus elaeagni (Del Guercio) 3 12

Capitophorus spp. 0 4

Hayhurstia atriplicis (Linnaeus) 0 4

Hyadaphis foeniculi (Passerini) 14 4

Landisaphis spp. 0 2

Lipaphis pseudobrassicae (Davis) 5 2

Macrosiphoniella spp. 0 1

Macrosiphum euphorbiae (Thomas) 2 6

Myzus persicae (Sulzer) 10 6

Pleotrichophorus spp. 0 1

Uroleucon spp. 2 4

Calaphidinae Panaphidini Therioaphis trifolii (Monell) f. maculata 32 84

Therioaphis spp. 1 0

Chaitophorinae Chaitophorini Chaitophorus spp. 1 1

Siphini Sipha flava (Forbes) 4 0

Drepanosiphinae — Drepanaphis acerifoliae (Thomas) 0 5

Drepanaphis spp. 2 1

Eriosomatinae Eriosomatini Colopha ulmicola (Fitch) 14 7

Tetraneura spp. 14 30

Fordini Asiphonella spp. 0 2

Pemphigini Pemphigus monophagus Maxson 1 0

Pemphigus populitransversus Riley 3 0

Pemphigus spp. 0 35

Saltusaphidinae Saltusaphidini Saltusaphis spp. 0 1

Unknown 1 10

Total 2933 966

Total – excluding Aphis gossypii 699 607

Table A1-3. Proportion of pumpkin leaf samples (n=20) infected with both PRSV and WMV, PRSV only, WMV only, or neither virus from each field.

Year Site County PRSV & WMV PRSV only WMV only No virus

2010 C Laporte 0.15 0 0.8 0.05

2010 G Laporte 0.05 0.05 0.55 0.35

2010 Z Laporte 0.2 0 0.3 0.5

2010 K White 0.85 0 0.15 0

2010 T20 Tipton 0 0.05 0.35 0.6

2010 T3 Tipton 0 0.2 0.05 0

2010 T5 Hamilton 0.15 0.15 0.1 0.6

2010 R Hancock 0 0 0.5 0.5

2010 S Parke 0 0 0.9 0.1

2010 E Daviess 0.9 0 0.1 0

2010 MC Sullivan 0.15 0 0.8 0.05

2010 MO Knox 0.05 0 0.9 0.05

2010 N Knox 0.75 0 0.25 0

2011 C Laporte 0 0 0.3 0.7

2011 G Laporte 0.4 0.05 0.3 0.25

2011 X Laporte 0.55 0.05 0.35 0.05

2011 MT Jasper 0.1 0.15 0.05 0.7

2011 K White 0 0 0.8 0.2

2011 O Tippecanoe 0 0 0.5 0.5

2011 M Tippecanoe 0 0 0.05 0.95

2011 SO Clinton 0 0 0.8 0.2

2011 T5 Hamilton 0.4 0 0.6 0

2011 SC Hamilton 0 0.35 0 0.65

2011 R Hancock 0 0.35 0.25 0.4

2011 B Hendricks 0.25 0.15 0.05 0.55

2011 L Putnam 0.25 0.15 0.4 0.2

2011 P Montgomery 0 0 0.9 0.1

2011 S Parke 0.1 0 0.8 0.1

2011 H Vermillion 0.6 0.1 0 0.2

Table A1-4. Relationship of PRSV presence/absence in fields and species explaining >95% of aphid community variation between presence/absence groups, analyzed with logistic regression and t-test: A) 2010, B) 2011. A) 2010:

Species Regression formula Residual SD z-value P t-test

Aphis gossypii y=1.56+0.00069x 11.02 0.32 >0.1 t=-0.63; df=6.56; P>0.1

Rhopalosiphum padi y=1.51+0.0051x 11.09 0.26 >0.1 t=-0.34; df=2.04; P>0.1

Aphis glycines y=1.85-0.032x 7.84 0.88 >0.1 t=0.88; df=1.01; P>0.1

Therioaphis trifolii y=1.85-0.032x 11.11 -0.24 >0.1 t=0.244; df=1.54; P>0.1

Rhopalosiphum y=1.83-0.030x 11.13 -0.19 >0.1 t=0.29; df=2.99; maidis P>0.1

Aphis craccivora y=0.38+0.52x 8.52 1.07 >0.1 t=-2.59; df=5.27; P<0.05

Hyadaphis foeniculi y=1.49+0.27x 10.72 0.45 >0.1 t=-1.18; df=11.00; P>0.1

Tetraneura spp. y=1.53+0.14x 11.07 0.28 >0.1 t=-0.34; df=1.82; P>0.1

B) 2011:

Species Regression formula Residual SD z-value P t-test

Aphis gossypii y=1.92-0.051x 14.23 -1.74 0.08 t=2.27; df=5.35; P<0.1

Aphis craccivora y=0.20+0.020x 20.26 0.45 >0.1 t=-0.99; df=9.20; P>0.1

Therioaphis trifolii y=0.31x-1.68 14.90 1.92 0.06 t=-3.121; df=13.67; P<0.01

Tetraneura spp. y=0.69-0.054x 20.84 -0.57 >0.1 t=0.47; df=6.45; P>0.1

Pemphigus spp. y=0.74-0.069x 21.01 -0.39 >0.1 t=0.35; df=8.93; P>0.1

Rhopalosiphum padi y=0.79-0.11x 20.97 -0.45 >0.1 t=0.43; df=10.77; P>0.1

Rhopalosiphum y=1.42-0.51x 19.68 -1.14 >0.1 t=1.04; df=7.21; maidis P>0.1

Aphis nerii y=0.40+0.074x 21.09 0.27 >0.1 t=-0.27; df=12.15; P>0.1

Capitophorus y=0.99-0.43x 19.70 -1.16 >0.1 t=1.26; df=12.55; elaeagni P>0.1

Anoecia spp. y=0.14+0.47x 19.99 0.99 >0.1 t=-1.21; df=13.24; P>0.1

Colopha ulmicola y=0.43+0.10x 21.09 0.28 >0.1 t=-0.30; df=;13.93 P>0.1

Aphis fabae y=0.76-0.32x 20.58 -0.75 >0.1 t=0.77; df=12.11; P>0.1

Aphis glycines y=0.76-0.28x 20.60 -0.75 >0.1 t=0.70; df=9.30; P>0.1

Uroleucon spp. y=1.20-27.43x 14.05 -0.004 >0.1 t=0.94; df=5.00; P>0.1

Acyrthosiphon pisum y=0.37+0.21 20.64 0.60 >0.1 t=-0.79; df=11.41; P>0.1

Table A1-5. Model selection proportion of PRSV-infected samples (arcsine square root transformed) by aphid community based on AICc (Akaike Information Criterion, corrected for small sample size) : A) 2010, B) 2011.

A) 2010:

Delta Log- AICc Likelihood Best Models AICc AICcWt

1. Tetraneura spp., Schizaphis graminum, Aphis glycines 18.66 0.00 0.43 -5.00

2. Tetraneura spp., Therioaphis trifolii, Myzus persicae, 19.23 0.57 0.32 -3.11 Aphis glycines

3. Tetraneura spp., Therioaphis trifolii, Rhopalosiphum 20.66 2.00 0.91 -1.04 maidis, Anoecia spp., Protaphis middletonii

4. Rhopalosiphum padi, Lipaphis pseudobrassicae, Colopha 21.87 3.20 0.09 2.07 ulmicola, Aphis gossypii, Aphis craccivora

B) 2011:

Delta Log- AICc Likelihood Best Models AICc AICcWt

1. Rhopalosiphum maidis, Hyadaphis foeniculi, Aphis -15.96 0.00 0.77 26.26 gossypii, Aphis glycines, Aphis craccivora, Aphis fabae

2. Aphis glycines, Aphis craccivora, Aphis fabae, Myzus -13.52 2.44 0.23 38.76 persicae, Hyadaphis foeniculi, Hayhurstia atriplicis, Pemphigus spp., Anoecia spp.

Table A1-6. Relationship of PRSV and aphid species from the best subsets equation best explaining the variation in proportion of samples inoculated, analyzed with linear regression: A) 2010, B) 2011.

A) 2010:

Species Regression formula Residual SE R2 F P

Tetraneura spp. y=0.52-0.012x 0.43 0.0031 0.034 >0.1

Schizaphis graminum y=0.44+0.066x 0.41 0.078 0.93 >0.1

Aphis glycines y=0.58-0.027x 0.39 0.18 2.37 >0.1

B) 2011:

Species Regression formula Residual SE R2 F P

Rhopalosiphum maidis y=0.62-0.11x 0.35 0.16 2.70 >0.1

Hyadaphis foeniculi y=0.49-0.30x 0.36 0.11 1.77 >0.1

Aphis gossypii y=0.59-0.0053x 0.32 0.32 6.67 <0.05

Aphis craccivora y=0.39+0.0016x 0.38 0.055 0.81 >0.1

Aphis fabae y=0.46-0.046x 0.38 0.025 0.35 >0.1

Table A1-7. Model selection proportion of WMV-infected samples (arcsine square root transformed) by aphid community based on AICc: A) 2010, B) 2011.

A) 2010:

Delta Log- AICc Likelihood Best Models AICc AICcWt

1. Rhopalosiphum padi, Aphis glycines, Acyrthosiphon -2.52 0.00 0.35 14.26 pisum, Aphis fabae

2. Therioaphis trifolii, Rhopalosiphum maidis, Myzus -1.79 0.73 0.24 19.10 persicae, Colopha ulmicola, Aphis gossypii

3. Rhopalosiphum padi, Aphis fabae, Acyrthosiphon -1.45 1.07 0.21 10.01 pisum

4. Schizaphis graminum, Rhopalosiphum maidis, Myzus -0.93 1.60 0.16 26.46 persicae, Colopha ulmicola, Aphis gossypii, Aphis fabae

5. Sipha flava, Rhopalosiphum maidis, Pemphigus spp., 2.11 4.63 0.03 37.95 Colopha ulmicola, Capitophorus elaeagni, Anoecia spp., Aphis gossypii

B) 2011:

Delta Log- AICc Likelihood Best Models AICc AICcWt

1. Rhopalosiphum rufiabdominalis, Rhopalosiphum 12.38 0.00 0.87 25.81 padi, Macrosiphum euphorbiae, Hyadaphis foeniculi, Anoecia spp., Aphis nerii, Aphis lacinariae, Acyrthosiphon pisum

2. Uroleucon spp., Pemphigus spp., Hyadaphis 18.39 6.01 0.04 1.47 foeniculi, Aphis gossypii

3. Aphis nerii, Aphis gossypii 18.89 6.51 0.03 -3.63

4. Uroleucon spp., Aphis nerii, Aphis gossypii 19.82 7.44 0.02 -1.91

5. Aphis nerii 19.93 7.55 0.02 -5.97

6. Rhopalosiphum maidis, Pemphigus spp., Hyadaphis 22.27 9.89 0.01 2.86 foeniculi, Aphis fabae, Aphis gossypii

Table A1-8. Relationship of WMV and aphid species from the best subsets equation best explaining the variation in proportion of samples inoculated, analyzed with linear regression: A) 2010, B) 2011.

A) 2010:

Species Regression formula Residual SE R2 F P

Rhopalosiphum padi y=1.35-0.0070x 0.34 0.47 9.70 <0.01

Aphis glycines y=1.04+0.0078x 0.47 0.012 0.14 >0.1

Acyrthosiphon pisum y=1.06+0.018x 0.47 0.0014 0.015 >0.1

Aphis fabae y=1.09-0.042x 0.46 0.023 0.26 >0.1

B) 2011:

Species

Regression formula Residual SE R2 F P

Rhopalosiphum rufiabdominalis y=0.84+0.016x 0.44 0.00067 0.0096 >0.1

Rhopalosiphum padi y=0.78+0.027x 0.44 0.018 0.26 >0.1

Macrosiphum euphorbiae y=0.86-0.059x 0.44 0.0088 0.12 >0.1

Hyadaphis foeniculi y=0.82+0.11x 0.44 0.011 0.16 >0.1

Anoecia spp. y=0.81+0.041x 0.44 0.018 0.26 >0.1

Aphis nerii y=0.67+0.11x 0.38 0.28 5.41 <0.05

Aphis lacinariae y=0.83+0.051x 0.44 0.0044 0.062 >0.1

Acyrthosiphon pisum y=0.83+0.015x 0.44 0.0069 0.097 >0.1

Table A1-9. Relationship of PRSV or WMV and abundance of total aphids, colonizers (Aphis gossypii), noncolonizers, and additional aphid species present early in sampling season (weeks 1 and 2), analyzed with linear regression. Aphid species present in two or fewer fields in the first two weeks were omitted from the latter analysis, including 2011 Colonizers: A) 2010, B) 2011.

A) 2010:

Species Virus Regression formula Residual SE R2 F P

Aphis craccivora PRSV y=0.53-0.032x 0.43 0.012 0.13 >0.1

WMV y=1.08-0.022x 0.47 0.0049 0.054 >0.1

Aphis glycines PRSV y=0.60-0.30x 0.40 0.12 1.54 >0.1

WMV y=1.11-0.16x 0.46 0.031 0.35 >0.1

Rhopalosiphum maidis PRSV y=0.45+0.077x 0.42 0.060 0.70 >0.1

WMV y=1.03+0.045x 0.46 0.017 0.19 >0.1

Rhopalosiphum padi PRSV y=0.64-0.16x 0.38 0.22 3.06 >0.1

WMV y=1.29-0.26x 0.33 0.51 11.36 <0.01

Therioaphis trifolii PRSV y=0.60-0.073x 0.40 0.12 1.45 >0.1

WMV y=1.10-0.025x 0.47 0.011 0.13 >0.1

Total Aphids PRSV y=0.58-0.014x 0.42 0.057 0.66 >0.1

WMV y=1.13-0.012x 0.46 0.035 0.40 >0.1

Colonizers PRSV y=0.54-0.028x 0.42 0.027 0.30 >0.1

WMV y=1.08-0.015x 0.47 0.0066 0.073 >0.1

Noncolonizers PRSV y=0.59-0.019x 0.42 0.059 0.69 >0.1

WMV y=1.14-0.018x 0.46 0.045 0.52 >0.1

B) 2011:

Species Virus Regression formula Residual SE R2 F P

Aphis craccivora PRSV y=0.41+0.0064x 0.39 0.00081 0.011 >0.1

WMV y=0.88-0.018x 0.44 0.0049 0.069 >0.1

Aphis fabae PRSV y=0.42+0.036x 0.39 0.0015 0.021 >0.1

WMV y=0.83+0.096x 0.44 0.0082 0.12 >0.1

Anoecia spp. PRSV y=0.36+0.18x 0.37 0.060 0.90 >0.1

WMV y=0.78+0.16x 0.43 0.033 0.49 >0.1

Colopha ulmicola PRSV y=0.46-0.11x 0.38 0.018 0.26 >0.1

WMV y=0.80+0.17x 0.44 0.031 0.44 >0.1

Pemphigus spp. PRSV y=0.80+0.17x 0.44 0.031 0.44 >0.1

WMV y=0.78+0.086x 0.43 0.042 0.61 >0.1

Rhopalosiphum padi PRSV y=0.41+0.051x 0.38 0.0069 0.097 >0.1

WMV y=0.85-0.013x 0.44 0.00035 0.0050 >0.1

Tetraneura spp. PRSV y=0.52-0.20x 0.35 0.16 2.63 >0.1

WMV y=0.81+0.083x 0.44 0.020 0.29 >0.1

Therioaphis trifolii PRSV y=0.31+0.076x 0.35 0.16 2.74 >0.1

WMV y=0.72+0.080x 0.41 0.14 2.30 >0.1

Total Aphids PRSV y=0.38+0.0069x 0.39 0.0047 0.066 >0.1

WMV y=0.56+0.041x 0.41 0.13 2.05 >0.1

Noncolonizers PRSV y=0.39+0.0055x 0.39 0.0029 0.041 >0.1

WMV y=0.56+0.042x 0.41 0.13 2.03 >0.1

Table A1-10. Relationship of PRSV presence/absence in fields and abundance of total aphids, colonizers (Aphis gossypii), noncolonizers, and additional aphid species present early in sampling season (weeks 1 and 2), analyzed with logistic regression and t-test. Aphid species present in two or fewer fields in the first two weeks were omitted from the latter analysis, including 2011 Colonizers: A) 2010, B) 2011.

A) 2010:

Species Regression formula Residual SD z-value P t-test

Aphis y=1.10+17.65x 9.00 0.004 >0.1 t=-1.99; df=10; craccivora P<0.1

Aphis glycines y=2.08-0.98x 10.78 -0.63 >0.1 t=0.44; df=1.16; P>0.1

Rhopalosiphum y=1.61+0.16x 11.10 0.23 0.1 t=-0.35; df=2.85; maidis P>0.1

Rhopalosiphum y=2.70-0.82x 9.24 -1.31 >0.1 t=0.68; df=1.03; padi P>0.1

Therioaphis y=2.00-0.19x 10.88 -0.55 >0.1 t=0.70; df=1.87; trifolii P>0.1

Total Aphids y=1.67+0.0069x 11.16 0.06 >0.1 t=-0.076; df=2.04; P>0.1

Colonizers y=1.25+16.69x 9.53 0.003 >0.1 t=-1.87; df=10; P<0.1

Noncolonizers y=1.89-0.037x 11.09 -0.28 >0.1 t=0.29; df=1.53; P>0.1

B) 2011:

Species Regression formula Residual SD z-value P t-test

Aphis craccivora y=0.38+0.066x 21.13 0.20 >0.1 t=-0.21; df=13.98; P>0.1

Aphis fabae y=0.47+0.22x 21.14 0.17 >0.1 t=-0.16; df=10.96; P>0.1

Anoecia spp. y=0.41+0.29x 21.10 0.27 >0.1 t=-0.25; df=10.67; P>0.1

Colopha ulmicola y=0.69-0.69x 20.82 -0.59 >0.1 t=0.53; df=9; P>0.1

Pemphigus spp. y=1.11-0.082x 18.89 -1.42 >0.1 t=1.40; df=8.17; P>0.1

Rhopalosiphum padi y=0.54-0.097x 21.16 -0.11 >0.1 t=0.091; df=7.15; P>0.1

Tetraneura spp. y=1.14-1.37x 18.11 -1.56 >0.1 t=1.72; df=9.21; P>0.1

Therioaphis trifolii y=-0.25+0.64x 18.48 1.25 >0.1 t=-1.77; df=12.14; P>0.1

Total Aphids y=0.72-0.031x 21.12 -0.22 >0.1 t=0.24; df=13.51; P>0.1

Noncolonizers y=0.78-0.040x 21.10 -0.27 >0.1 t=0.30; df=13.57; P>0.1

Figure A1-1. Location of fields monitored throughout Indiana in 2010 (n=13) and 2011 (n=16): =2010, =2011, =2010 & 2011.

Figure A1-2. Relationship between on-plant aphid counts and pan trap melon aphid counts in fields.

Figure A1-3. Relationship between proportion (prop) of PRSV- and WMV-inoculated samples and total aphid alightment (aphid-days, the average alightment count trap-1 in a field each week, summed across the sampling season), analyzed by linear regression on arcsine-square root transformed proportion data. (A) 2010 prop PRSV and total aphids. (B) 2011 prop PRSV and total aphids. (C) 2010 prop WMV and total aphids. (D) 2011 prop WMV and total aphids.

Figure A1-4. Relationship between PRSV presence in fields and total aphid alightment (aphid-days the average alightment count trap-1 in a field each week, summed across the sampling season). Logistic regressions (panels A and C): =PRSV presence (1) or absence (0) in fields by aphid alightment, = fitted regression values of estimated PRSV odds probability given aphid alightment in a field. Boxplots (panels B and D): aphid alightment in fields grouped by PRSV presence or absence in fields, displaying median values and interquartile ranges, and pairwise group comparisons tested with student’s t test; =outliers. (A) 2010 logistic regression of PRSV presence probability and total aphids. (B) 2010 median total aphid alightment by PRSV presence or absence. (C) 2011 logistic regression of PRSV presence probability and total aphids. (D) 2011 median total aphid alightment by PRSV presence or absence.

Figure A1-5. Relationship between PRSV presence in fields and early season (abundance data from week 1+ week 2) colonizer or cowpea aphid alightment in fields are presented as in Figure A1-4. (A) 2010 logistic regression of PRSV presence probability and colonizers. (B) 2010 median colonizer alightment by PRSV presence or absence. (C) 2010 logistic regression of PRSV presence probability and cowpea aphid. (D) 2010 median cowpea aphid alightment by PRSV presence or absence.

Appendix B: Chapter 2 Supplemental Material

Table B2-1. Outer model bootstrap validation results after 600 iterations on the four PLS-path models investigating the relationships between landscape composition at varied spatial scales and/or weed coverage with aphid alightment and PRSV or WMV infection within fields in 2010 or 2011. Values presented include original coefficient estimate (Est.), bias (distance of bootstrap mean from original estimate), standard deviation (SD), and 5% and 95% confidence intervals. Only blocks with multiple manifest variables are included. Land cover blocks with largest effect on virus infection in bold type. Manifest Variables Block Est. Bias SE 5% 95% Aphid Row Crops -7.78 7.39 57.60 -137.00 29.38 Land Alightment Forest/Shrub -3.17 3.46 26.50 -35.60 20.02 Cover 2010 Pasture/Hay/Grass -5.07 5.10 36.80 -64.50 18.56 (1 km) Urban -0.96 1.39 6.97 -17.80 10.28 Row Crops -2.20 9.19 23.20 -71.50 32.74 Land Forest/Shrub -1.09 5.36 15.50 -95.00 18.14 Cover Pasture/Hay/Grass -1.09 4.27 9.56 -27.40 13.65 (2 km) Urban 0.55 1.84 4.47 -9.00 9.81 Row Crops -10.70 8.96 28.10 -190.00 7.47 Land Forest/Shrub -7.40 6.05 20.70 -128.00 9.87 Cover Pasture/Hay/Grass -4.69 4.10 11.30 -75.80 2.74 (3 km) Urban -1.24 1.59 6.55 -29.50 2.84 Row Crops -15.50 7.70 23.40 -91.40 12.96 Land Forest/Shrub -12.29 5.69 19.90 -83.40 14.05 Cover Pasture/Hay/Grass -4.75 2.89 7.86 -45.30 2.17 (4 km) Urban -1.64 0.42 4.05 -12.20 3.82 Row Crops -16.17 9.65 32.60 -159.00 11.59 Land Forest/Shrub -13.28 7.54 27.00 -127.00 10.89 Cover Pasture/Hay/Grass -3.64 2.37 9.14 -40.30 3.35 (5 km) Urban -1.55 1.16 3.49 -13.60 1.86 Aphid Row Crops 3.42 7.01 66.30 -164.93 106.04 Land Alightment Forest/Shrub 3.15 6.40 51.70 -127.80 91.30 Cover 2011 Pasture/Hay/Grass 2.81 3.56 38.10 -89.44 58.84 (1 km) Urban 0.61 1.56 18.20 -48.07 29.81 Row Crops -4.78 -0.17 18.60 -107.75 6.21 Land Forest/Shrub -3.11 0.16 13.30 -72.98 4.49 Cover Pasture/Hay/Grass -2.49 -0.58 12.60 -71.61 4.12 (2 km) Urban -0.45 0.38 6.90 -40.99 1.92 Row Crops -2.00 -8.66 34.60 -173.64 3.26 Land Forest/Shrub -0.69 -5.96 23.00 -104.46 3.57 Cover Pasture/Hay/Grass -0.11 -5.72 21.70 -98.49 2.98 (3 km) Urban -0.45 -2.89 11.40 -61.86 1.16 Row Crops -0.81 -0.12 12.00 -74.59 5.41 Land Forest/Shrub 0.28 -0.25 6.82 -16.57 12.64 Cover Pasture/Hay/Grass -0.11 -0.35 7.65 -37.00 5.05 (4 km) Urban 0.23 -0.19 4.65 -28.21 2.63 Row Crops -0.18 -0.92 12.50 -44.18 14.92 Land Forest/Shrub 0.70 -0.76 6.03 -11.50 14.10 Cover Pasture/Hay/Grass 0.25 -0.83 7.64 -18.89 13.40 (5 km) Urban 0.43 -0.32 5.22 -18.45 5.24

Table B2-2. Inner model bootstrap validation results after 600 iterations on the four PLS-path models investigating the relationships between landscape composition at varied spatial scales and /or weed coverage with aphid alightment and PRSV or WMV infection within fields in 2010 or 2011. Values presented include original coefficient estimate (Est.), bias (distance of bootstrap mean from original estimate), standard deviation (SD), and 5% and 95% confidence intervals.

Exogenous Endogenous Variable Variable Est. Bias SD 5% 95% Aphid 1 km Total Noncol. -0.46 -4.00 109.00 -1470.00 1.97 Alightment 2 km Total Noncol. 1.44 -6.40 78.40 -405.00 81.42 2010 *3 km Total Noncol. -1.67 -6.02 195.00 -4760.00 -0.34 4 km Total Noncol. 0.70 -17.65 541.00 -0.41 1588.20 5 km Total Noncol. -0.71 21.12 678.00 -2000.00 -0.64 *Weed Cover Total Noncol. -1.02 9.70 135.00 -20.90 -0.64 Total Noncol. PRSV 0.58 0.02 0.11 -593.00 0.73 Total Noncol. WMV -0.40 0.00 0.29 -86.20 0.16 Aphid 1 km Total Noncol. -0.18 0.17 1.01 -3.52 1.22 Alightment 2 km Total Noncol. -0.76 0.69 1.28 -5.03 0.58 2011 *3 km Total Noncol. 1.39 -1.20 1.97 0.02 10.79 4 km Total Noncol. -1.13 1.00 2.94 -16.61 0.85 5 km Total Noncol. -0.13 -0.20 2.32 -4.82 5.56 Weed Cover Total Noncol. -0.02 0.00 0.30 -0.63 0.55 Total Noncol. PRSV -0.01 0.01 0.27 -0.49 0.51 Total Noncol. WMV 0.37 -0.03 0.25 0.22 0.73

Table B2-3. Proportion of WMV- or PRSV-infected samples and total stems (2010) or total cover m-2 (2011) of WMV or PRSV reservoir species within and surrounding pumpkin fields in 2010 (n=10) and 2011 (n=16).

A) 2010:

Weed Species Regression rse F P PRSV Chenopodium album y=0.40-0.01x 0.37 0.11 >0.05 WMV Amaranthus spp. y=0.82+0.03x 0.44 0.04 >0.05 Chenopodium album y=0.79+0.03x 0.44 0.28 >0.05 Digitaria sanguinalis y=0.81+0.05x 0.44 0.20 >0.05 Ipomoea hederacea y=0.65+0.04x 0.39 2.60 >0.05 Mollugo verticillata y=0.66+0.06x 0.43 0.71 >0.05 Solanum ptycanthum y=0.93-0.03x 0.44 0.32 >0.05 1 km soy y=0.81+0.13x 0.44 0.01 >0.05 2 km soy y=1.11-0.95x 0.44 0.27 >0.05 3 km soy y=1.37-1.89x 0.42 0.90 >0.05 4 km soy y=1.06-0.73x 0.44 0.13 >0.05 5 km soy y=1.14-1.02x 0.44 0.26 >0.05 1 km alfalfa y=0.88-22.92x 0.44 0.10 >0.05 2 km alfalfa y=0.97-37.28x 0.43 0.66 >0.05 3 km alfalfa y=0.98-45.44x 0.42 0.82 >0.05 4 km alfalfa y=0.97-52.94x 0.43 0.69 >0.05 5 km alfalfa y=0.93-29.44x 0.44 0.18 >0.05

B) 2011:

Weed Species Regression rse F P PRSV Chenopodium album y=0.31+10.15.x 0.34 4.26 >0.05 WMV Amaranthus spp. y=0.78+1.82.x 0.43 0.81 >0.05 Chenopodium album y=0.78+4.91.x 0.43 0.61 >0.05 Digitaria sanguinalis y=0.80+10.00.x 0.43 0.69 >0.05 Ipomoea hederacea y=0.91-3.01.x 0.43 0.98 >0.05 Mollugo verticillata y=0.81+1.00.x 0.44 0.25 >0.05 Solanum ptycanthum y=0.79+4.06.x 0.44 0.46 >0.05 Trifolium spp. y=0.81+47.51.x 0.44 0.44 >0.05 1 km soy y=1.21-1.34x 0.41 2.14 >0.05 2 km soy y=0.97-0.51x 0.44 0.12 >0.05 3 km soy y=0.99-0.56x 0.44 0.14 >0.05 4 km soy y=0.55+1.15x 0.44 0.39 >0.05 5 km soy y=0.51+1.27x 0.44 0.45 >0.05 1 km alfalfa y=0.79+16.43x 0.44 0.43 >0.05 2 km alfalfa y=0.82+4.21x 0.44 0.03 >0.05 3 km alfalfa y=0.77+13.88x 0.44 0.19 >0.05 4 km alfalfa y=0.78+15.01x 0.44 0.15 >0.05 5 km alfalfa y=0.76+19.24x 0.44 0.22 >0.05

Table B2-4. Arcsine-square root transformed proportion of PRSV-infected samples and cover field-1 of WMV or PRSV interactions between aphid trap catch and total weed cover/total stems, and reservoir species (Chenopodium album) within pumpkin fields in 2010 (n=10) and 2011 (n=16).

A) 2010:

2 Factors β1(se) t1 β2(se) t2 β3(se) t3 R (adj.) F Chenopodium 0.14(0.19) 0.73 0.08(0.06) 1.29 -0.02(0.02) -0.81 -0.15 0.60 album x aphids total stems x 0.12(0.15) 0.79 0.44(0.36) 1.23 -0.02(0.02) -1.19 0.11 1.36 aphids

B) 2011:

2 Factors β1(se) t1 β2(se) t2 β3(se) t3 R (adj.) F Chenopodium 43.56(23.10) 1.89 -0.01(0.01) -0.76 -0.78(0.61) -1.28 0.30 3.02 album x aphids total cover x -1.27(1.63) -0.78 -0.02(0.02) -0.80 -0.80(0.65) -1.22 -0.13 0.46 aphids

Table B2-5. Arcsine-square root transformed proportion of WMV-infected samples and cover field-1 of WMV or PRSV interactions between aphid trap catch and total weed cover, and reservoir species (total res. weed cover, soy, alfalfa) within and surrounding pumpkin fields in 2010 (n=10) and 2011 (n=16). Surrounding cover regressed at scale best predicting aphid alightment, determined by PLS-PMs.

A) 2010

2 Factors β1(se) t1 β2(se) t2 β3(se) t3 R (adj.) F total res. stems 0.01(0.06) 0.15 -0.07(0.13) -0.56 0.00(0.01) 0.50 0.11 1.36 x aphids total stems x 0.06(0.19) 0.33 0.24(0.50) 0.49 -0.02(0.02) -0.72 -0.04 0.88 aphids soy (3 km) x 1.30 0.16 0.03 0.12 -0.28 -0.31 -0.20 0.50 aphids alfalfa (3 km) x -777.86 -1.31 -0.25 -1.59 124.17 1.23 0.13 1.46 aphids

B) 2011

2 Factors β1(se) t1 β2(se) t2 β3(se) t3 R (adj.) F total res. -0.79(4.02) -0.20 0.01(0.03) 0.20 0.05(0.15) 0.34 -0.07 0.70 cover x aphids total cover x 1.95(1.70) 1.15 0.03(0.02) 1.60 -0.06(0.08) -0.76 0.10 1.52 aphids soy (3km) x -1.05 -0.13 0.01 0.15 0.02 0.07 -0.09 0.63 aphids alfalfa (3 km) -101.30 -0.71 0.00 -0.15 5.39 0.84 -0.01 0.96 x aphids

Appendix C: Chapter 3 Supplemental Material

Ac111034

Ac30228

Ac42054

Ac46423

Figure C3-1. Nonneutral alleles identified by LOSITAN as undergoing balancing selection, which were subsequently removed from the dataset.

Table C3-1. Aphis craccivora polymorphism, heterozygosity, and equilibrium across 22 loci among populations, with full dataset containing repeat MLGs.

Ac1 Ac1 Ac1 Ac1 Ac1 Ac Ac Ac Ac Ac Ac Ac Ac Ac Ac Ac Ac Ac Ac Ac Ac Ac 000 010 022 053 070 199 240 247 259 314 330 330 438 441 455 699 807 827 882 916 94 949 Pop 49 26 92 72 71 23 56 66 86 85 20 58 65 39 40 31 98 46 99 05 71 89 0.09 0.09 0.09 0.09 0.18 0.0 0.2 0.3 0.0 0.0 0.0 0.0 0.0 0.1 0.2 0.1 0.0 0.2 0.1 0.4 0.4 0.0 K Ho 1 1 1 1 2 91 73 64 91 00 91 91 91 82 73 00 91 73 82 55 55 91 0.08 0.08 0.08 0.08 0.16 0.0 0.3 0.4 0.0 0.4 0.0 0.0 0.0 0.4 0.5 0.0 0.0 0.4 0.1 0.3 0.4 0.0 He 7 7 7 7 5 87 51 96 87 80 87 87 87 96 00 95 87 83 65 51 34 87 ------0.04 0.04 0.04 0.04 0.10 0.0 0.2 0.2 0.0 1.0 0.0 0.0 0.0 0.6 0.4 0.0 0.0 0.4 0.1 0.2 0.0 0.0 Fis 8 8 8 8 0 48 24 67 48 00 48 48 48 33 55 53 48 36 00 94 48 48 0.13 0.13 0.13 0.13 0.13 0.1 0.4 0.6 0.1 0.1 0.1 0.1 0.1 0.1 0.2 0.1 0.1 0.4 0.3 0.4 0.3 0.1 M Ho 3 3 3 3 3 33 00 43 33 33 33 33 33 33 67 33 33 67 33 00 33 33 0.12 0.12 0.12 0.12 0.12 0.1 0.3 0.4 0.1 0.4 0.1 0.1 0.1 0.4 0.4 0.1 0.1 0.5 0.2 0.4 0.4 0.1 He 4 4 4 4 4 24 20 77 24 44 24 24 24 44 80 24 24 00 78 98 20 24 ------0.07 0.07 0.07 0.07 0.07 0.0 0.2 0.3 0.0 0.7 0.0 0.0 0.0 0.7 0.4 0.0 0.0 0.0 0.2 0.1 0.2 0.0 Fis 1 1 1 1 1 71 50 48 71 00 71 71 71 00 44 71 71 67 00 96 06 71 0.40 0.40 0.40 0.40 0.40 0.4 0.4 0.6 0.4 0.4 0.4 0.4 0.4 0.4 0.6 0.4 0.4 0.8 0.4 0.8 0.8 0.6 O Ho 0 0 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 0.32 0.32 0.32 0.32 0.32 0.3 0.3 0.5 0.3 0.3 0.3 0.3 0.3 0.3 0.5 0.3 0.3 0.4 0.3 0.4 0.4 0.4 He 0 0 0 0 0 20 20 00 20 20 20 20 20 20 00 20 20 80 20 80 80 20 ------0.25 0.25 0.25 0.25 0.25 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.6 0.2 0.6 0.6 0.4 Fis 0 0 0 0 0 50 50 00 50 50 50 50 50 50 00 50 50 67 50 67 67 29 0.22 0.33 0.33 0.22 0.22 0.2 0.3 0.2 0.2 0.2 0.2 0.2 0.2 0.3 0.5 0.2 0.2 0.2 0.2 0.6 0.2 0.3 Mt Ho 2 3 3 2 2 22 33 22 22 22 22 22 22 33 56 22 22 22 22 67 22 33 0.19 0.27 0.27 0.19 0.19 0.1 0.2 0.4 0.1 0.4 0.1 0.1 0.1 0.5 0.4 0.1 0.1 0.4 0.1 0.4 0.4 0.2 He 8 8 8 8 8 98 78 94 98 94 98 98 98 00 75 98 98 44 98 44 94 78 ------0.12 0.20 0.20 0.12 0.12 0.1 0.2 0.5 0.1 0.5 0.1 0.1 0.1 0.3 0.1 0.1 0.1 0.5 0.1 0.5 0.5 0.2 Fis 5 0 0 5 5 25 00 50 25 50 25 25 25 33 69 25 25 00 25 00 50 00 1.00 1.00 1.00 1.00 1.00 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 IL Ho 0 0 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 0.50 0.50 0.50 0.50 0.50 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 He 0 0 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 97

------1.00 1.00 1.00 1.00 1.00 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 Fis 0 0 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 0.94 0.95 0.95 0.95 0.95 0.9 1.0 1.0 0.9 0.9 0.9 0.9 0.9 0.9 0.9 1.0 0.9 1.0 1.0 0.9 1.0 0.9 WI Ho 7 0 0 0 0 50 00 00 50 50 47 50 50 50 50 00 50 00 00 50 00 50 0.49 0.49 0.49 0.49 0.49 0.4 0.5 0.5 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.5 0.4 0.5 0.5 0.4 0.5 0.4 He 9 9 9 9 9 99 00 00 99 99 99 99 99 99 99 00 99 00 00 99 00 99 ------0.90 0.90 0.90 0.90 0.90 0.9 1.0 1.0 0.9 0.9 0.9 0.9 0.9 0.9 0.9 1.0 0.9 1.0 1.0 0.9 1.0 0.9 Fis 0 5 5 5 5 05 00 00 05 05 00 05 05 05 05 00 05 00 00 05 00 05 1.00 1.00 1.00 1.00 1.00 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 MI Ho 0 0 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 0.50 0.50 0.50 0.50 0.50 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 He 0 0 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ------1.00 1.00 1.00 1.00 1.00 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 Fis 0 0 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 1.00 1.00 1.00 1.00 1.00 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 sIN Ho 0 0 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 0.50 0.50 0.50 0.50 0.50 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 He 0 0 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ------1.00 1.00 1.00 1.00 1.00 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 Fis 0 0 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 wcI 0.95 0.94 0.94 0.94 0.95 0.9 0.9 0.9 0.9 0.9 0.9 0.9 0.9 0.9 0.9 0.9 0.9 0.9 0.9 0.9 0.9 0.9 N Ho 0 7 7 7 0 50 50 50 50 50 50 50 47 50 47 50 50 50 47 50 50 50 0.49 0.49 0.49 0.49 0.49 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 He 9 9 9 9 9 99 99 99 99 99 99 99 99 99 99 99 99 99 99 99 99 99 ------0.90 0.90 0.90 0.90 0.90 0.9 0.9 0.9 0.9 0.9 0.9 0.9 0.9 0.9 0.9 0.9 0.9 0.9 0.9 0.9 0.9 0.9 Fis 5 0 0 0 5 05 05 05 05 05 05 05 00 05 00 05 05 05 00 05 05 05 nea rM 1.00 1.00 1.00 1.00 1.00 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 T Ho 0 0 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 0.50 0.50 0.50 0.50 0.50 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 He 0 0 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00

Fis ------98

1.00 1.00 1.00 1.00 1.00 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 0 0 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 nea 0.90 0.94 0.95 0.90 0.90 0.8 0.9 0.9 0.9 0.9 0.9 0.9 0.9 0.9 0.9 0.9 0.9 0.9 0.9 1.0 0.9 0.9 rM Ho 0 7 0 0 0 95 47 00 00 00 00 00 00 00 50 00 00 50 00 00 00 47 0.49 0.49 0.49 0.49 0.49 0.4 0.4 0.5 0.4 0.5 0.4 0.4 0.4 0.5 0.4 0.4 0.4 0.4 0.4 0.5 0.5 0.4 He 5 9 9 5 5 94 99 00 95 00 95 95 95 00 99 95 95 99 95 00 00 99 ------0.81 0.90 0.90 0.81 0.81 0.8 0.9 0.8 0.8 0.8 0.8 0.8 0.8 0.8 0.9 0.8 0.8 0.9 0.8 1.0 0.8 0.9 Fis 8 0 5 8 8 10 00 00 18 00 18 18 18 00 05 18 18 05 18 00 00 00 nea rO K1 1.00 1.00 1.00 1.00 1.00 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 0 Ho 0 0 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 0.50 0.50 0.50 0.50 0.50 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 He 0 0 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ------1.00 1.00 1.00 1.00 1.00 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 Fis 0 0 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 nea rO 0.83 0.83 0.83 0.83 0.83 0.8 0.8 0.8 0.8 0.8 0.8 0.8 0.8 0.8 1.0 0.8 0.8 0.8 0.8 0.8 0.8 1.0 K8 Ho 3 3 3 3 3 33 33 33 33 33 33 33 24 33 00 33 33 33 33 33 33 00 0.48 0.48 0.48 0.48 0.48 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.5 0.4 0.4 0.4 0.4 0.4 0.4 0.5 He 6 6 6 6 6 86 86 86 86 86 86 86 84 86 00 86 86 86 86 86 86 00 ------0.71 0.71 0.71 0.71 0.71 0.7 0.7 0.7 0.7 0.7 0.7 0.7 0.7 0.7 1.0 0.7 0.7 0.7 0.7 0.7 0.7 1.0 Fis 4 4 4 4 4 14 14 14 14 14 14 14 00 14 00 14 14 14 14 14 14 00 btw nM 1.00 1.00 1.00 1.00 1.00 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 tK Ho 0 0 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 0.50 0.50 0.50 0.50 0.50 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 He 0 0 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ------1.00 1.00 1.00 1.00 1.00 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 Fis 0 0 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 MI 0.00 0.00 0.00 0.00 0.00 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 bl Ho 0 0 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00

He 0.00 0.00 0.00 0.00 0.00 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 99

0 0 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00

Fis – – – – – – – – – – – – – – – – – – – – – – 0.76 0.76 0.82 0.76 0.87 0.7 0.8 0.8 0.7 0.7 0.7 0.7 0.7 0.8 0.8 0.7 0.7 0.9 0.7 0.8 0.8 0.7 OH Ho 5 5 4 5 5 50 82 24 65 50 50 65 50 24 82 65 65 41 65 24 82 65 0.47 0.47 0.48 0.47 0.49 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 He 2 2 4 2 2 69 93 84 72 69 69 72 69 84 93 72 72 98 72 84 93 72 ------0.61 0.61 0.70 0.61 0.77 0.6 0.7 0.7 0.6 0.6 0.6 0.6 0.6 0.7 0.7 0.6 0.6 0.8 0.6 0.7 0.7 0.6 Fis 9 9 0 9 8 00 89 00 19 00 00 19 00 00 89 19 19 89 19 00 89 19 0.00 0.00 0.00 0.00 0.00 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 KY Ho 0 0 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 0.00 0.00 0.00 0.00 0.00 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 He 0 0 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00

Fis – – – – – – – – – – – – – – – – – – – – – – wcI 0.00 0.00 0.00 0.00 0.00 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Nbl Ho 0 0 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 0.00 0.00 0.00 0.00 0.00 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 He 0 0 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00

Fis – – – – – – – – – – – – – – – – – – – – – – nea rM Tbl 0.00 0.00 0.00 0.00 0.00 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 3 Ho 0 0 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 0.00 0.00 0.00 0.00 0.00 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 He 0 0 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00

Fis – – – – – – – – – – – – – – – – – – – – – – Car olC 0.00 0.00 0.00 0.00 0.00 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 o Ho 0 0 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 0.00 0.00 0.00 0.00 0.00 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 He 0 0 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00

Fis – – – – – – – – – – – – – – – – – – – – – – nea rO 0.00 0.00 0.00 0.00 0.00 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 K Ho 0 0 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00

100

0.00 0.00 0.00 0.00 0.00 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 He 0 0 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00

Fis – – – – – – – – – – – – – – – – – – – – – – btw nM 0.00 0.00 0.00 0.06 0.00 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 O Ho 0 0 0 3 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 0.00 0.00 0.00 0.06 0.00 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 He 0 0 0 1 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 - 0.03 Fis – – – 2 – – – – – – – – – – – – – – – – – –

Table C3-2. Aphis craccivora polymorphism, heterozygosity, and equilibrium across 22 loci among populations, after excluding the effect of clonality by removing all repeat MLGs. Ac Ac Ac1 Ac1 Ac1 Ac Ac Ac Ac Ac Ac Ac Ac Ac Ac Ac Ac Ac Ac Ac Ac Ac 100 101 022 053 070 199 240 247 259 314 330 330 438 441 455 699 807 827 882 916 94 949 Pop 049 026 92 72 71 23 56 66 86 85 20 58 65 39 40 31 98 46 99 05 71 89 0.1 0.1 0.12 0.12 0.25 0.1 0.3 0.5 0.1 0.0 0.1 0.1 0.1 0.2 0.3 0.1 0.1 0.3 0.2 0.6 0.6 0.1 K Ho 25 25 5 5 0 25 75 00 25 00 25 25 25 50 75 25 25 75 50 25 25 25 0.1 0.1 0.11 0.11 0.21 0.1 0.4 0.4 0.1 0.2 0.1 0.1 0.1 0.3 0.4 0.1 0.1 0.4 0.2 0.4 0.4 0.1 He 17 17 7 7 9 17 30 69 17 45 17 17 17 75 30 17 17 92 19 30 92 17 ------0.0 0.0 0.06 0.06 0.14 0.0 0.1 0.0 0.0 1.0 0.0 0.0 0.0 0.3 0.1 0.0 0.0 0.2 0.1 0.4 0.2 0.0 Fis 67 67 7 7 3 67 27 67 67 00 67 67 67 33 27 67 67 38 43 55 70 67 0.1 0.1 0.12 0.12 0.12 0.1 0.3 0.7 0.1 0.1 0.1 0.1 0.1 0.1 0.3 0.1 0.1 0.6 0.2 0.3 0.3 0.1 M Ho 25 25 5 5 5 25 75 14 25 25 25 25 25 25 75 25 25 25 50 75 75 25 0.1 0.1 0.11 0.11 0.11 0.1 0.3 0.5 0.1 0.3 0.1 0.1 0.1 0.3 0.4 0.1 0.1 0.4 0.2 0.4 0.4 0.1 He 17 17 7 7 7 17 05 00 17 05 17 17 17 05 30 17 17 92 19 92 92 17 ------0.0 0.0 0.06 0.06 0.06 0.0 0.2 0.4 0.0 0.5 0.0 0.0 0.0 0.5 0.1 0.0 0.0 0.2 0.1 0.2 0.2 0.0 Fis 67 67 7 7 7 67 31 29 67 90 67 67 67 90 27 67 67 70 43 38 38 67 0.2 0.2 0.25 0.25 0.25 0.2 0.2 0.5 0.2 0.2 0.2 0.2 0.2 0.2 0.5 0.2 0.2 0.7 0.2 0.7 0.7 0.5 O Ho 50 50 0 0 0 50 50 00 50 50 50 50 50 50 00 50 50 50 50 50 50 00 0.2 0.2 0.21 0.21 0.21 0.2 0.2 0.5 0.2 0.2 0.2 0.2 0.2 0.2 0.5 0.2 0.2 0.4 0.2 0.4 0.4 0.3 He 19 19 9 9 9 19 19 00 19 19 19 19 19 19 00 19 19 69 19 69 69 75 ------0.0 ------0.0 ------F 0.1 0.1 0.14 0.14 0.14 0.1 0.1 00 0.1 0.1 0.1 0.1 0.1 0.1 00 0.1 0.1 0.6 0.1 0.6 0.6 0.3

is 101

43 43 3 3 3 43 43 43 43 43 43 43 43 43 43 00 43 00 00 33 0.1 0.2 0.28 0.14 0.14 0.1 0.2 0.1 0.1 0.1 0.1 0.1 0.1 0.2 0.5 0.1 0.1 0.1 0.1 0.7 0.1 0.2 Mt Ho 43 86 6 3 3 43 86 43 43 43 43 43 43 86 71 43 43 43 43 14 43 86 0.1 0.2 0.24 0.13 0.13 0.1 0.2 0.4 0.1 0.4 0.1 0.1 0.1 0.4 0.4 0.1 0.1 0.4 0.1 0.4 0.4 0.2 He 33 45 5 3 3 33 45 59 33 59 33 33 33 90 90 33 33 59 33 59 59 45 ------0.0 0.1 0.16 0.07 0.07 0.0 0.1 0.6 0.0 0.6 0.0 0.0 0.0 0.4 0.1 0.0 0.0 0.6 0.0 0.5 0.6 0.1 Fis 77 67 7 7 7 77 67 89 77 89 77 77 77 17 67 77 77 89 77 56 89 67 1.0 1.0 1.00 1.00 1.00 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 IL Ho 00 00 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 0.5 0.5 0.50 0.50 0.50 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 He 00 00 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ------1.0 1.0 1.00 1.00 1.00 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 Fis 00 00 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 0.5 0.5 0.50 0.50 0.50 0.5 1.0 1.0 0.5 0.5 0.5 0.5 0.5 0.5 0.5 1.0 0.5 1.0 1.0 0.5 1.0 0.5 WI Ho 00 00 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 0.3 0.3 0.37 0.37 0.37 0.3 0.5 0.5 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.5 0.3 0.5 0.5 0.3 0.5 0.3 He 75 75 5 5 5 75 00 00 75 75 75 75 75 75 75 00 75 00 00 75 00 75 ------0.3 0.3 0.33 0.33 0.33 0.3 1.0 1.0 0.3 0.3 0.3 0.3 0.3 0.3 0.3 1.0 0.3 1.0 1.0 0.3 1.0 0.3 Fis 33 33 3 3 3 33 00 00 33 33 33 33 33 33 33 00 33 00 00 33 00 33 1.0 1.0 1.00 1.00 1.00 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 MI Ho 00 00 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 0.5 0.5 0.50 0.50 0.50 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 He 00 00 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ------1.0 1.0 1.00 1.00 1.00 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 Fis 00 00 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 1.0 1.0 1.00 1.00 1.00 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 sIN Ho 00 00 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 0.5 0.5 0.50 0.50 0.50 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 He 00 00 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ------1.0 1.0 1.00 1.00 1.00 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 Fis 00 00 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00

102

wcI 0.5 0.5 0.50 0.50 0.50 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.0 0.5 0.5 0.5 N Ho 00 00 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 0.3 0.3 0.37 0.37 0.37 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.0 0.3 0.3 0.3 He 75 75 5 5 5 75 75 75 75 75 75 75 75 75 75 75 75 75 00 75 75 75 ------0.3 0.3 0.33 0.33 0.33 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3 Fis 33 33 3 3 3 33 33 33 33 33 33 33 33 33 33 33 33 33 – 33 33 33 nea rM 1.0 1.0 1.00 1.00 1.00 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 T Ho 00 00 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 0.5 0.5 0.50 0.50 0.50 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 He 00 00 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ------1.0 1.0 1.00 1.00 1.00 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 Fis 00 00 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 nea 0.3 0.6 0.66 0.33 0.33 0.3 0.5 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.6 0.3 0.3 0.6 0.3 1.0 0.3 0.6 rM Ho 33 67 7 3 3 33 00 33 33 33 33 33 33 33 67 33 33 67 33 00 33 67 0.2 0.4 0.44 0.27 0.27 0.2 0.3 0.5 0.2 0.5 0.2 0.2 0.2 0.5 0.4 0.2 0.2 0.4 0.2 0.5 0.5 0.4 He 78 44 4 8 8 78 75 00 78 00 78 78 78 00 44 78 78 44 78 00 00 44 ------0.2 0.5 0.50 0.20 0.20 0.2 0.3 0.3 0.2 0.3 0.2 0.2 0.2 0.3 0.5 0.2 0.2 0.5 0.2 1.0 0.3 0.5 Fis 00 00 0 0 0 00 33 33 00 33 00 00 00 33 00 00 00 00 00 00 33 00 nea rO K1 1.0 1.0 1.00 1.00 1.00 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 0 Ho 00 00 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 0.5 0.5 0.50 0.50 0.50 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 He 00 00 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ------1.0 1.0 1.00 1.00 1.00 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 Fis 00 00 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 nea rO 0.5 0.5 0.50 0.50 0.50 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 1.0 0.5 0.5 0.5 0.5 0.5 0.5 1.0 K8 Ho 00 00 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 0.3 0.3 0.37 0.37 0.37 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.5 0.3 0.3 0.3 0.3 0.3 0.3 0.5 He 75 75 5 5 5 75 75 75 75 75 75 75 75 75 00 75 75 75 75 75 75 00

Fis ------103

0.3 0.3 0.33 0.33 0.33 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3 1.0 0.3 0.3 0.3 0.3 0.3 0.3 1.0 33 33 3 3 3 33 33 33 33 33 33 33 33 33 00 33 33 33 33 33 33 00 btw nM 1.0 1.0 1.00 1.00 1.00 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 tK Ho 00 00 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 0.5 0.5 0.50 0.50 0.50 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 He 00 00 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ------1.0 1.0 1.00 1.00 1.00 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 Fis 00 00 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 MI 0.0 0.0 0.00 0.00 0.00 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 bl Ho 00 00 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 0.0 0.0 0.00 0.00 0.00 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 He 00 00 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00

Fis – – – – – – – – – – – – – – – – – – – – – – 0.2 0.2 0.40 0.20 0.60 0.2 0.6 0.4 0.2 0.2 0.2 0.2 0.2 0.4 0.6 0.2 0.2 0.8 0.2 0.4 0.6 0.2 OH Ho 00 00 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 0.1 0.1 0.32 0.18 0.42 0.1 0.4 0.3 0.1 0.1 0.1 0.1 0.1 0.3 0.4 0.1 0.1 0.4 0.1 0.3 0.4 0.1 He 80 80 0 0 0 80 20 20 80 80 80 80 80 20 20 80 80 80 80 20 20 80 ------0.1 0.1 0.25 0.11 0.42 0.1 0.4 0.2 0.1 0.1 0.1 0.1 0.1 0.2 0.4 0.1 0.1 0.6 0.1 0.2 0.4 0.1 Fis 11 11 0 1 9 11 29 50 11 11 11 11 11 50 29 11 11 67 11 50 29 11 0.0 0.0 0.00 0.00 0.00 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 KY Ho 00 00 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 0.0 0.0 0.00 0.00 0.00 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 He 00 00 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00

Fis – – – – – – – – – – – – – – – – – – – – – – wcI 0.0 0.0 0.00 0.00 0.00 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Nbl Ho 00 00 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 0.0 0.0 0.00 0.00 0.00 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 He 00 00 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00

Fis – – – – – – – – – – – – – – – – – – – – – – nea rM 0.0 0.0 0.00 0.00 0.00 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 T3 Ho 00 00 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00

He 0.0 0.0 0.00 0.00 0.00 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 104

00 00 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00

Fis – – – – – – – – – – – – – – – – – – – – – – Car olC 0.0 0.0 0.00 0.00 0.00 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 o Ho 00 00 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 0.0 0.0 0.00 0.00 0.00 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 He 00 00 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00

Fis – – – – – – – – – – – – – – – – – – – – – – nea rO 0.0 0.0 0.00 0.00 0.00 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 K Ho 00 00 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 0.0 0.0 0.00 0.00 0.00 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 He 00 00 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00

Fis – – – – – – – – – – – – – – – – – – – – – – btw nM 0.0 0.0 0.00 0.50 0.00 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 O Ho 00 00 0 0 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 0.0 0.0 0.00 0.37 0.00 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 He 00 00 0 5 0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 - 0.33 Fis – – – 3 – – – – – – – – – – – – – – – – – –

105

VITA

106

VITA

Gina M. Angelella Department of Entomology, Purdue University 901 W. State St., West Lafayette, IN 47907 [email protected] (765)496-1717 EDUCATION Doctor of Philosophy in Entomology Expected Fall 2015 Purdue University Advisor, Dr. Ian Kaplan Dissertation title, Tracking plant virus infections through multiple dimensions: A search for sources of nonpersistent virus vectors and reservoirs at local and regional scales. Master of Science in Entomology Fall 2008 University of Georgia Advisor, Dr. David G. Riley Thesis title, Pine pollen effects on Frankliniella occidentalis and Frankliniella fusca (Thysanoptera: Thripidae) reproduction. Bachelor of Science in Environmental Biology/Zoology Spring 2004 Michigan State University

GRANTS AND AWARDS 2015 USDA NIFA AFRI Predoctoral Fellowship: Is the behavior of crop virus-vectors influenced by facultative endosymbionts? ($34,500) 2015 J.T. Eaton and Company Scholarship, Purdue University ($1,000) 2014 Bilsland Dissertation Fellowship, Purdue University ($9,000) 2014 Indiana Academy of Science Senior Research Grant: Do aphid secondary endosymbionts mediate plant virus transmission? ($2,414) 2014 BASF Professional Pest Control Scholarship, Purdue University ($1,000)

107

2012 IVGA Grant to Support Virus Research, Indiana Vegetable Growers Association: A search for weedy sources of virus infection in Indiana cucurbits ($500) 2011 Second Place (President’s Prize), student poster competition, P-IE Section 5, Annual Meeting of the Entomological Society of America, Reno, NV 2010 Frederick N. Andrews Fellowship, Purdue University ($36,000) 2007 Tobacco Education and Research Council Grant, George Kennedy et al. Predicting and managing Tomato spotted wilt virus in tobacco ($439,258 total; $221,525 UGA subcontract; $51,000 UGA entomology portion—Riley and Angelella).

PUBLICATIONS Peer-Reviewed: Angelella, G.M., D.S. Egel, J.D. Holland, C. E. Williams, and I. Kaplan. 2015. Differential life history trait associations of aphids with nonpersistent viruses in cucurbits. Environmental Entomology 44(3):562–573. Riley, D.G., G.M. Angelella, and R.M. McPherson. 2011. Pine pollen dehiscence relative to thrips population dynamics. Entomologia Experimentalis et Applicata 138(3):223–233. Kaplan I., Angelella G., Blubaugh C., Braasch J., Caceres V., Prado J., Quesada C., Sadof C.S., Spigler M., Thompson S. 2011. Book review: Relationships of natural enemies and non-prey foods. American Entomologist 57:116-117. Angelella, G.M., and D.G. Riley. 2010. Effects of pine pollen supplementation in an onion diet on Frankliniella fusca reproduction. Environmental Entomology 39(2):505–512.

Non-Refereed: Angelella, G. 2013. Aphids land less often in weedy fields. Vol. 2, Issue 2: The Pumpkin News (ed. by K. Ziehm), Greenwich, NY. Angelella, G. 2012. Aphids can be a virus source. Vol. 1, Issue 7: The Pumpkin News (ed. by K. Ziehm), Greenwich, NY. Angelella, G., and R. Foster. 2011. Aphids in late season pumpkins. No. 544: Vegetable Crops Hotline (ed. by D. Egel), Vincennes, IN. Angelella, G. 2010. Do soybean aphids spread viruses in pumpkins? No. 531: Vegetable Crops Hotline (ed. by D. Egel), Vincennes, IN.

108

PRESENTATIONS Do facultative endosymbionts alter stylet-born virus transmission by aphids? Angelella, G.M., P. Nachappa, V. Nalam, J. White, and I. Kaplan. Entomological Society of America Annual Meeting, Portland, OR, Nov. 2014 Tracking plant virus infections through multiple dimensions: A search for sources of nonpersistent virus vectors and reservoirs at local and regional scales. Angelella, G.M., and I. Kaplan. Entomological Society of America Annual Meeting, Austin, TX, Nov. 2013 (Invited Talk). Local weed communities vs. landscape composition as drivers of aphid alightment in crops. Angelella, G.M., J. Holland, and I. Kaplan. Indiana Academy of Science Annual Meeting, Indianapolis, IN, Mar. 2013. Examining the role of aphid dynamics in pumpkin virus infections. Angelella, G.M., and I. Kaplan. Great Lakes Vegetable Working Group, Lafayette, IN, Feb. 2013. Aphid colonization patterns in pumpkins. Angelella, G.M., and I. Kaplan. Illiana Vegetable Growers Symposium, Schererville, IN, Jan. 2013 (Invited Talk). Local weed communities vs. landscape composition as drivers of aphid alightment in crops. Angelella, G.M., J. Holland, and I. Kaplan. Entomological Society of America Annual Meeting, Knoxville, TN, Nov. 2012. The phenology of aphid alatae as potential vectors of non-persistent virus in Midwestern pumpkins. Angelella, G.M., and I. Kaplan. Entomological Society of America Annual Meeting, Reno, NV, Nov. 2011. (Poster presentation) Surveying for aphids in Indiana cucurbits. Angelella, G. M., and I. Kaplan. Indiana Horticultural Congress, Indianapolis, IN, Jan. 2011 (Invited Talk). Pine pollen effects on Frankliniella occidentalis and F. fusca reproduction. Angelella, G.M., and D. Riley. Entomological Society of America Annual Meeting, Reno, NV, Nov. 2008. Studies on thrips reproduction. Angelella, G.M., D. Riley, and R. McPherson. Georgia-Florida Tobacco Tour, Tifton, GA, Jun. 2008. The potential effects of pollen on thrips reproduction. Angelella, G.M., and D. Riley. Georgia Entomological Society Annual Meeting, Cordele, GA, Apr. 2008. Bioassay methods for measuring thrips reproduction. Angelella, G.M., and D. Riley. Entomological Society of America Annual Meeting, San Diego, CA, Dec. 2007. (Poster presentation)

109

Effects of pollen on thrips reproduction. Angelella, G.M., and D. Riley. Georgia Entomological Society Annual Meeting, Athens, GA, Apr. 2007. (Poster presentation)

RELEVANT_EMPLOYMENT  Environmental Educator, U.S. Peace Corps, Village Navur, Tavush Marz, Armenia, 2004–2006. Worked with Armenian counterpart to design and implement environmental education curricula in area village schools, while increasing Armenian language environmental education resources available to the villages; organized afterschool environmental activities, English language lessons, running and cooking clubs, and an ecological summer camp for area children; and authored successful Peace Corps SPA grant, funding Village Navur Secondary School gymnasium renovations.  Entomology Intern, All Taxa Biodiversity Index, Great Smokey Mountains National Park, TN, 2003. Aided field collections of aquatic macroinvertebrates, performed stream quality assessments, and identified invertebrate species collected.

TEACHING EXPERIENCE  Graduate Teaching Certification Program, Purdue University. Attended teaching workshops as well as classroom observations with feedback from students and experienced instructors, and guidance on the development of a teaching portfolio.  Introductory Genetics Lab, Agronomy Dept., Purdue University, 2013–2014. Delivered lectures, facilitated lab exercises and graded course materials.  Insect Ecology Lab, Entomology Dept., Purdue University, 2013. Prepared and delivered lecture, facilitated lab exercises and led class discussions for undergraduates.  Insect Natural History Lab, Entomology Dept., University of Georgia, 2007. Prepared and delivered introductory taxonomy lessons, facilitated field collected trips and graded insect collections.  Organismal Biology for Undergraduate Science Majors, Biology Dept., University of Georgia, 2006–2007. Facilitated writing-intensive laboratory lessons and experiments, designed lab assessment exams and graded projects and exams.  Attended the course, Pedagogy for WID (Writing in the Disciplines), University of Georgia, 2006. Introductory seminar to teaching writing intensive courses.

110

PROFESSIONAL SERVICE AND OUTREACH  Invited reviewer, Environmental Entomology, Economic Entomology.  Volunteer Judge, Purdue Undergraduate Research Symposium.  Volunteer for annual university insect outreach events, e.g., Bug Bowl, Purdue University; Insectival, the University of Georgia; the Georgia State Fair.  Student volunteer at Entomological Society of America annual meetings.

PROFESSIONAL AFFILIATIONS Entomological Society of America, Ecological Society of America, Indiana Academy of Sciences, International Society for Pest Information