Epidemiological Studies of Soybean Vein Necrosis Virus and Potential Resistance Mechanisms to Its Vector Neohydatothrips Variabi

Home , Frankliniella tritici, Necrosis, Neohydatothrips, Vitivirus

University of Arkansas, Fayetteville ScholarWorks@UARK

Theses and Dissertations

12-2018 Epidemiological Studies of Soybean Vein Necrosis Virus and Potential Resistance Mechanisms to its Vector Neohydatothrips variabilis (Beach) Jing Zhou University of Arkansas, Fayetteville

Follow this and additional works at: https://scholarworks.uark.edu/etd Part of the Epidemiology Commons, Molecular Biology Commons, and the Plant Pathology Commons

Recommended Citation Zhou, Jing, "Epidemiological Studies of Soybean Vein Necrosis Virus and Potential Resistance Mechanisms to its Vector Neohydatothrips variabilis (Beach)" (2018). Theses and Dissertations. 3095. https://scholarworks.uark.edu/etd/3095

This Dissertation is brought to you for free and open access by ScholarWorks@UARK. It has been accepted for inclusion in Theses and Dissertations by an authorized administrator of ScholarWorks@UARK. For more information, please contact [email protected], [email protected]. Epidemiological Studies of Soybean Vein Necrosis Virus and Potential Resistance Mechanisms to its Vector Neohydatothrips variabilis (Beach)

A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Plant Science

Jing Zhou Qingdao Agricultural University Bachelor of Science in Biotechnology, 2008 University of Arkansas Master of Science in Cell and Molecular Biology, 2012

December 2018 University of Arkansas

This dissertation is approved for recommendation to the Graduate Council.

______Ioannis E. Tzanetakis, Ph.D. Dissertation Director

______Pengyin Chen, Ph.D. Donn T. Johnson, Ph.D. Committee Member Committee Member

______Garry V. McDonald, Ph.D. John. C. Rupe, Ph.D. Committee Member Committee Member

ABSTRACT

Soybean (Glycine max (L.) Merrill) is one the most important crops in global agriculture with annual production of over 260 million metric tons. As the dependence of a growing global population to soybean has increased, so does the importance of soybean diseases and pests. Over

200 pathogens attack soybean; among them, viruses pose a major threat to the soybean industries accounting for approximately 10% of the annual yield reduction caused by diseases in the past two decades. Soybean vein necrosis virus (SVNV) is a relatively newly discovered virus causing the homonymous disease. The widespread occurrence of the disease in major soybean producing regions in North America and its negative impact on seed quality led to the work presented here; research that aims to better understand virus epidemiology so as to develop effective virus control and disease management tools. In order to further understand the potential roles of weeds in SVNV cycle, surveys were conducted to determine the presence of SVNV among 32 weed species collected from soybean fields in Arkansas. Kudzu (Pueraria montana), a common weed present in millions of acres in Southeastern United States, can sustain SVNV replication in a systemic manner without developing virus-like symptoms. SVNV – a localized virus in soybean, could move systemically with the assistance of bean pod mottle virus, one of the most economically important soybean viruses. The ineffectiveness of pesticides in thrips management highlights the need to identify potential resistance mechanisms to the primary and highly efficient vector of SVNV – Neohydatothrips variabilis (Beach). The comparison of transmission efficiency of thrips fed with polypeptides containing RGD motif and N-linked glycosylation sites of SVNV glycoproteins indicated that blocking putative cellular receptors prior to virus acquisition could significantly reduce the virus transmission efficiency. Due to the lack of resistance to SVNV, efforts were made to identify genotypes with resistance to the vector, which

could modify vector behavior and reduce the incidence of transmission and disease. Screening of soybean accessions with differential leaf pubescence levels revealed that feeding damage caused by thrips differs among accessions and is weakly correlated to their pubescence levels.

ACKNOWLEDGEMENTS

I am deeply grateful to my major advisor Dr. Ioannis Tzanetakis for his guidance and continuous support through my PhD program. His sharp vision and timeless pursue of high- quality science deeply influenced me. The tremendous freedom he provided in his laboratory and his criticism combined with genuine concern through each critical step of my graduate study play a pivotal role in making me an independent researcher.

With a special mention to my committee member Dr. Donn Johnson, who provided valuable suggestions regarding the research and dissertation, I am sincerely thankful to him and all other committee members Dr. Pengyin Chen, Dr. John Rupe and Dr. Garry McDonald for their great effort in helping me growing as a better scientist along the way.

The complexity of my PhD projects called for intellectual input from experts beyond my committee members, it is impossible to accomplish the work I presented here without their help.

I would like to give special thanks to Dr. Mariusz Lewandowski at Warsaw University of Life

Sciences, Dr. Suresh Thallapuranam and Dr. Roger Koeppe at University of Arkansas, Dr. Anna

Whitfield at North Carolina State University and Dr. Seed Sabanadzovic at Mississippi State

University for their heartwarming support. I also greatly appreciate Dr. Ken Korth, Mr. Scott

Winters, Ms. Jessica Kivett and Mr. Bob Holland for their generous assistance which made the

challenging projects went smoothly.

Last but not the least, I would like to thank all current and past members of the plant

virology laboratory as well as faculties, staffs and students of the Department of Plant Pathology

during my stay at the University of Arkansas. I greatly appreciate everyone’s support, help and

encouragement which motivated me to be a better person.

DEDICATION

This dissertation is dedicated to my dear father for his unfailing love and unconditional support. His love is my inspiration which motivates me in all aspects of my life. Without his guidance and patience, I could not reach where I am today.

TABLE OF CONTENTS

Chapter I …………………………………………………………………………………………..1 Introduction ...………….………………………………………………………………………2 Genome Organization ……………………………………………………………………...….3 Symptomology and Host Range …………………………………………………………..…..6 Disease Diagnosis …………………………………………………………………………..…9 Transmission ……………………………………………………………………………...….10 Disease Management …………………………………………………………………….…..12 Summary ……………………………………………………………………………………..14 References ……………………………………………………………………………………22

Chapter II ………………………………………………………………………………………..28 Abstract ………………………………………………………………………………………29 Introduction …………………………………………………………………………………..30 Materials and Methods ……………………………………………………………………….31 Field survey …………………………………………………………………………….…31 Greenhouse studies ………………………………………………………………………..32 Alternative virus vectors ………………………………………………………………….35 Results ………………………………………………………………………………………..35 Field Survey ………………………………………………………………………………35 Greenhouse study …………………………………………………………………………36 Alternative virus vector …………………………………………………………………...36 Discussion ………………………………………………………………………………...….37 References …………………………………………………………………………...……….43

Chapter III ……………………………………………………………………………………….46 Abstract ………………………………………………………………………………………47 Introduction …………………………………………………………………………………..48 Materials and Methods …………………………………………………………………….…49 Plant material and virus isolates …………………………………………………………..49 Virus inoculation ………………………………………………………………………….50

Nucleic acid extraction and reverse transcription ………………………………………...50 PCR ……………………………………………………………………………………….51 Dot blot immunoassay ………………………………………………………………..…..51 Results ……………………………………………………………………………………….52 ‘Hutcheson’ is a restricted host for SVNV ……………………………………………….52 ‘Hutcheson’ is a permissive host for BPMV and SMV …………………………………..53 SVNV moves systemically in ‘Hutcheson’ in the presence of BPMV but not SMV .....…53 Discussion ……………………………………………………………………………………55 References …………………………………………………………………………………....65

Chapter IV ……………………………………………………………………………………….69 Abstract …..…………………………………………………………………………………..70 Introduction ………………………………………………………………………………...... 71 Materials and Methods ……………………………………………………………………….74 Neohydatothrips variabilis rearing …..……………………………………………...…….74 Peptides design and synthesis ..……………………………………………………….…..74 Peptide delivery ………………………………………………………..………………….75 Peptide effects on transmission ...…………………………………………………………76 Peptide sequencing ...……………………………………………………………………...76 Results ………………………………………………………………………………………..77 Peptide delivery …………………………………………………………………………...77 Transmission ………………………………………………………………...……………77 Statistical analysis …………………………………………………………………….…..78 Peptide sequencing ………………………………………………………………………..78 Discussion ……………………………………………………………………………………78 References ……………………………………………………………………………..……..92

Chapter V ………………………………………………………………………………………..97 Abstract ………………………………………………………………………………………98 Introduction ………………………………………………………………………………...... 99 Materials and Methods ………………………………………………………………..…….101

Plant materials ………………………………………………………………………...…101 Neohydatothrips variabilis colony and virus inoculation …………………………...…..101 Screening for thrips – resistant/tolerant accessions ………………………………….….101 Data analysis ………………………………………………………………………….....102 Results …………………………………………………………………………………...….103 Discussion ……………………………………………………………………………..……104 References ……………………………………………………………………………..……116

Conclusions …………………………………………………………………………………….120 References …………………………………………………………………………………..122

PUBLISHED PAPERS

Portions of Chapter II were previously published as: Zhou, J., Aboughanem-Sabanadzovic, N., Sabanadzovic, S. and Tzanetakis, I. E. 2018. First report of soybean vein necrosis virus infecting kudzu (Pueraria montana) in the United States of America. Plant Disease. 102: 1674.

UNPUBLISHED PAPERS

Chapter I Zhou, J. and Tzanetakis, I. E. Soybean Vein Necrosis Virus: an Emerging Virus in North America. Virus Genes. Accepted.

Chapter I

Soybean Vein Necrosis Virus: an Emerging Virus in North America

Introduction

Soybean vein necrosis virus (SVNV) is an orthotospovirus naturally infecting soybean.

The virus is transmitted primarily by the soybean thrips (Neohydatothrips variabilis, Beach) in a persistent and propagative manner and causes localized infections on soybean leaves. As a distinct member of the genus Orthotospovirus, Family Tospoviridae, SVNV shares minimal similarity with all established species in the genus and represents a new clade in the genus evolution (Zhou et al., 2011; Zhou and Tzanetakis, 2013).

SVNV was first reported in Arkansas and Tennessee, U. S. in 2008 associated with symptoms that initiated with vein clearing followed by lesions and necrosis (Tzanetakis et al.,

2009). In subsequent years, symptoms were observed in the two aforementioned states but also several other soybean-growing areas including Illinois, Wisconsin, Michigan, Ohio,

Pennsylvania, Delaware, Kansas, Oklahoma, Kentucky, Missouri, Mississippi, Louisiana and

Alabama (Zhou et al., 2011; Zhou, 2012; Ali and Abdalla, 2013; Conner et al., 2013; Escalante et al., 2018; Han et al., 2013; Jacobs and Chilvers., 2013; Kleczewski, 2016; Smith et al., 2013).

Nowadays, the virus has been confirmed in at least 22 states across the U.S. as well as Canada and Egypt (Abd El-Wahab and El-Shazly, 2017) and vein necrosis has become the most prevalent virus disease in North America (Zhou and Tzanetakis, 2013).

The rapid spread of SVNV raised concerns about its economic impact; providing impetus to gain a better understanding of the fundamental aspects of the virus and the disease, and to develop appropriate control strategies. This review highlights our knowledge on the biology and epidemiology of the virus as well as diagnostics and control strategies for the disease.

Genome Organization

The genome of SVNV has a typical orthotospovirus organization which consists of three

single-stranded RNA segments that are designated as L-, M- and S- RNAs. The pleomorphic virions of orthotopoviruses range in size from 80-120 nm. The polymerase and nucleoprotein of orthotospoviruses are enclosed within a host-derived lipid membrane with the two viral glycoproteins – Gn and Gc projecting from the surface (Whitfield et al., 2005; Bag et al., 2015).

In silico analysis of the SVNV genome revealed classic features of members in the Tospoviridae, such as all three RNA segments have the highly conserved 5’ terminal sequence (AGAGCA1–6)

predicted to be critical in replication and transcription signal (Sherwood et al., 2000). Many

other, atypical, attributes of the genome will be discussed here in more detail.

SVNV L RNA is 9010 nucleotides (nt) in length and contains a single open reading

frame (ORF) in the negative orientation. The 19 nt of 5’- and 3’- ends are complementary to each other putatively leading to the circularization of the molecule forming a panhandle structure,

similar to other orthotospoviruses (Sherwood et al., 2000). The region between nt 8980-185

codes for a polyprotein of 336 KDa with five motifs (A (DxxKWS 539–544), B (QGxxxxxSS 527–

535), C (SDD 665-667), D (K712) and E (EFxSE 721–725)) present in RNA-dependent RNA

polymerases (RdRp). Those motifs alongside motif F (Kx451–452, KxQR459–462 and TxxDRxIY463–

470), present only in some orthotospoviruses, are part of the “U-shape” crevice formed by typical

RdRp domains (Roberts et al., 1995; Bruenn, 2003) The RNA and the polyprotein it encodes have distinct properties when compared with other members of the genus including size - the longest one alongside bean necrotic mosaic virus (BNeMV), the closest-related virus to SVNV.

Additionally, a Lysine-rich extension (TSSSGSK2900–2906 and

KWSKPKKKKKPKAKPKKSKKKHNK2908–2931) with unknown function is identified at the

carboxy-terminal of the protein (Zhou et al., 2011; de Oliveira et al., 2011).

The M RNA has 4955 nt with the first and last 27 showing almost perfect complementarity and potentially forming a panhandle structure. This RNA codes two ORFs which are separated by a 267 nt A/U-rich intergenic region (IGR). ORF158-1008, codes for a 35

KDa non-structural protein (NSm). The presence of highly conserved LxDx40G motif of the 30K

movement protein superfamily suggests that the protein is involved in cell-to-cell movement

(Mushegian and Koonin, 1993; Melcher, 2000; Silva et al., 2001), however, the Leu residue has

been substituted by an Ile at the beginning of the motif. The ‘‘P/D-L-X motif’’ and

phospholipase A2 catalytic sites, present in some orthotospovirus orthologs such as tomato

spotted virus (TSWV) and groundnut bud necrosis virus (GBNV) are absent from the SVNV

counterpart (Silva et al., 2001). ORF24863-1276 codes for the precursor of the virion glycoproteins

(Gn/Gc). Signal cleavage between Cys378 and Ser379 yields two proteins: Gn (43 KDa) and Gc

(91 KDa). The Gn protein contains several signature motifs present in orthotospovirus orthologs

including a RGD29-31 domain, which is crucial in virion - cell receptor attachment (Melcher,

2000; Silva et al., 2001) as well as several N-Glycosylation sites (N25, 229,343) and transmembrane

domains (aa6–28, 317–339, 349–371). The SVNV Gc protein has a series of highly conversed sequences

present in orthotopovirus orthologs including Lys702, a T-X-T714-716, CTGxC730–734 and

TSxWGCEExxCXAxxxGxxxGxC754–776 (Cortez et al., 2002) whereas N-Glycosylation sites

transmembrane domains are present at N5, 20, 171 and aa77-99, respectively. SVNV has the largest

glycoproteins among all the members in the genus with a long amino acid tail on the C –

terminus.

S RNA is 2603 nt long and contains two ORFs in opposite orientations. The first and last six nucleotides of this segment are complementary, similar to the other two RNAs. The untranslated region is highly structured with 5’- and 3’- UTRs being 58 and 70 nt long, respectively. The ORF159-1381 encodes NSs, a 50-kDa non-structural protein predicted to be an

RNAi suppressor (Takeda et al., 2004). Conserved GK178–179 and DExx148–151 comprise the

Walker A and B motifs which interact with ATP/ADP phosphates and coordinate/bind Mg2+ ions during ATP hydrolysis (Caruthers and McKay, 2002; Lokesh et al., 2010). The remaining

ORF2533-1700 codes for a putative nucleoprotein (N) of 31 KDa. The protein has an RNA binding

Lysine-rich motif KKDGKGKKSK264–273, as well as several discrete RNA-interacting amino acids (PSN7–9, RK51–52, RY54–55, and KK73–74), domains that probably allow nucleoprotein to participate in RNA synthesis together with the RNA L polyprotein as shown for members of the

Bunyavirales (Dunn et al., 1995; Flick and Pettersson, 2001; Flick et al., 2003; Kainz et al.,

2004; Kukkonen et al., 2005). The two ORFs are separated by a 318 nt A/U rich IGR, one the smallest among members in the genus (de Oliveira et al., 2012).

Phylogenetic analyses based on all coding regions of the genome indicate that SVNV and

BNeMV belong to a distinct clade that shares almost equidistance between American and

Eurasian lineages (Zhou et al., 2011; de Oliveira et al., 2011; de Oliveira et al., 2012; Chen et al.,

2013; Fig. 1). Serological relationship between SVNV and other orthotopoviruses species representing existing serogroups or distinct serotypes within the genus verified SVNV having a distinct serotype (Huang et al., 2017), corroborating with its unique phylogenetic placement. The genomic divergence of SVNV-BNeMV clade from the other orthotopovirus groups and the unique features discussed here suggest that SVNV is the type member of a novel evolutionary lineage of Fabaceae-infecting orthotopoviruses.

Symptomology and Host Range

The early symptoms of SVNV infection include clearing along the main veins,

sometimes with small light-green to yellow patches distributed between veins. Affected areas

expand and become chlorotic and eventually necrotic as leaves mature (Zhou et al., 2011;

Tzanetakis et al., 2009). The distinction between infected and non-infected areas may be blurring

on fully-developed leaflets in the field and sometimes could be mistaken for other disorders (Fig.

2A); such distinction becomes much clear as disease progresses (Fig. 2B). Unlike diseases such

as frogeye leaf spot or bacterial blight which also cause foliar lesions, SVNV infection-caused

lesions expand through leaf veins to the surrounding areas and are rarely surrounded with halos.

They are irregular-shaped and tend to unevenly distribute on leaf blades which probably mirror

the vector preferred feeding areas (Fig. 2C). Later in the season, lesions coalesce leading to

scotched appearance or leaf death (Fig. 2F). Symptom intensities seem to vary in both

greenhouse and field conditions (Zhou, personal observation). Mild symptoms exhibit as thread-

shaped lesions along the main vein or other irregular shapes of yellow patches which take up

minimal areas of the leaf blade (Fig. 2D); whereas more aggressive symptoms display as yellow,

or reddish-brown to dark brown lesions covering the major portion of the blade (Fig. 2E). Such

symptom variations may be correlated to different host genotypes as suggested by Anderson

(Chen et al., 2013) which could represent as tolerant versus susceptible cultivars to either the

virus or thrips feeding, or it could result from the fact that virus infection occurs at different

growth stages of soybeans which causes different levels of damage to the plant. The timing of

first appearance of disease symptoms in different soybean growing regions varies from June to

October (Hajimorad et al., 2015; Chitturi et al., 2018; Keough et al., 2018; Zhou, 2018); on the other hand, the timing also varies between years depending on weather patterns. In general

terms, in hotter and drier conditions disease emerges early in the season, possibly due to higher vector populations (Zhou, Tzanetakis, personal observations). Disease symptoms are usually first observed on the lower canopy moving upwards as newly emerged leaves are the preferential feeding sites for thrips (Chitturi et al., 2018; Zhou, 2018).

Apart from soybean, SVNV has been reported to naturally infect another leguminous crop - yard-long bean (Vigna unguiculata spp. sesquipedalis) (Escalante et al., 2018) Infected yard-long beans exhibit vein yellowing and chlorotic spots surrounded by necrosis on leaflets.

The virus is only detected on symptomatic but not asymptomatic samples; how the virus moves on the plant (locally or systemically), however, has not been determined yet (Valverde, personal communication). Several studies have been performed to investigate the role of indigenous weeds in disease epidemiology. So far, natural SVNV infection has been reported on ivyleaf morning glory (Ipomoea hederacea Jacq.), entireleaf morning glory (Ipomoea hederacea var. integriuscula) and kudzu (Pueraria montana) (Zhou and Tzanetakis, 2013; Zhou,

2018; Sikora et al., 2018). The high incidence of morning glory species in soybean fields and the fact that ivyleaf morning glory is a natural host of SVNV stimulated researchers to further explore the role of weeds in virus dissemination. Both ivyleaf morning glory and pitted morning glory (Ipomoea lacunose L.) sustain virus replication in green house experiment whereas a 3- year field survey in Alabama detected low infection rate of SVNV infection on entireleaf morning glory but not on pitted morning glory (Anderson 2017; Sikora et al., 2018). A recent study revealed kudzu – a weed species in the Fabaceae as an asymptomatic, systemic host of

SVNV (Zhou et al., 2018). Given that kudzu is a perennial weed which presents in millions of acres in Southeastern U.S. and has the overlapping geographic range with many major soybean producing states, it is possible that this plant species may serve as the major reservoir for SVNV;

providing overwintering or early season population growth habitats for viruliferous thrips prior

to moving to soybean.

All characterized SVNV isolates cause localized infections on soybean where the virus is

restricted in and around the clearing or lesion areas (Zhou and Tzanetakis, 2013; Hajimorad et

al., 2015; Khatabi et al., 2012). Symptoms observed in soybean fields have been reproduced in

green house studies using either mechanical or vector inoculation (Zhou and Tzanetakis, 2013;

Anderson 2017; Irizarry et al., 2018). Several plant species were tested as alternative hosts in

greenhouse studies, most of which produce local-lesions indicating hypersensitive reactions to

the virus. SVNV infection causes similar symptoms on legume species including cowpea (Vigna

unguiculate), mungbean (Vigna radiata), medicago (Medicago truncatula) and pigeon pea

(Cajanus cajan) (Fig. 3). Typical symptoms on these species include chlorotic lesions which

become either necrotic or coalesce resulting in senescence or even death of inoculated leaves.

Disease symptoms on non-leguminous hosts, however vary. On Nicotiana benthamiana, SVNV

produces necrosis on inoculated leaflets which expands to newly-emerged leaves and the

systemic movement of the virus leads to stem collapse and plant death (Zhou and Tzanetakis,

2013). On buckwheat (Fagopyrum esculentum), another systemic host, virus infection displays

chlorosis to necrosis whereas on melon (Cucumis melo), only small sunken gray lesions ranging

from were observed (Irizarry et al., 2018). In addition, few species including chrysanthemum

(Dendranthema grandiflorum) and pumpkin (Cucurbita pepo L.) were proved asymptomatic hosts (Zhou and Tzanetakis, 2013). According to Anderson (2017), Palmer amaranth

(Amaranthus palmeri S. Wats.) and Redroot pigweed (Amaranthus retroflexus L.), were tested positive in thrips inoculation experiment; however plants were not observed for a prolonged time period to determine whether they are symptomatic or asymptomatic hosts.

Disease Diagnosis

Typical disease symptoms caused by the infection of orthotospoviruses include chlorotic

lesions and necrosis, independent of localized or systemic hosts (Bag et al., 2015). In the case of

SVNV, the virus remains localized on soybean and exhibits symptoms as discussed above, which

could aid the diagnosis of the disease. Confirmation of virus infection on verified hosts and

diagnosis on new hosts, however, requires accurate and sensitive detection methods. There are

currently two types of assays routinely used for SVNV detection: immunological- and PCR- based. For immunodiagnostics, polyclonal antibodies were generated against the recombinant E. coli-expressed nucleocapsid protein of the virus (Khatabi et al., 2012) enabling to detect the virus using dot blot or enzyme-linked immunosorbent assay (ELISA). A dilution of 1: 2000 of this antiserum could detect SVNV in sap extract from naturally infected soybean leaf tissues diluted up to 1:512, according to Khatabi and co-workers. Apart from that, several ELISA kits are available in the market. Our previous studies showed DAS-ELISA using 1:200 dilution of polyclonal antibody generated against SVNV N protein from one commercial vendor is capable of detecting the virus from leaf tissues grinded in buffer at the ratio of 1/20 (w/v) but not in any further sap dilutions (Zhou, 2012). Likewise, other studies also showed inconsistent results when

ELISA was used for SVNV detection: the virus was detected using RT-PCR in plant tissues that were tested as negative with ELISA (Anderson, 2017; Irizarry, 2016). These results suggest the efficacy of immunological-based diagnosis varies; probably due to the quality of the antibodies used.

In terms of nucleic acid-based detection techniques, several PCR detection assays have been developed and successfully applied in virus diagnosis (Zhou et al., 2011; Zhou, 2012; Ali and Abdalla, 2013; Conner et al., 2013; Escalante et al., 2018; Han et al., 2013; Jacobs and

Chilvers., 2013; Kleczewski, 2016; Smith et al., 2013). Most assays were designed using the N

gene as the target for amplification given that 1) the population structure analysis based on N

protein of SVNV revealed a minimal diversity across a wide geographic area (Zhou and

Tzanetakis, 2013), which makes it ideal in designing assays that can detect even diverse virus

isolates; 2) N gene is highly expressed allowing for sensitive detection. In Zhou and Tzanetakis

(Zhou and Tzanetakis, 2013), the assay could detect the virus in 4 pg of RNA extracted from

naturally infected soybean tissues in 30 PCR cycles or 40 pg of the same RNA sample in 20

cycles. It is generally accepted that real-time PCR (qPCR) has higher sensitivity than conventional PCR in virus detection. However, the development of qPCR assays with lower detection limit than conventional PCR is challenging for SVNV. Only one qPCR assay has been published for SVNV by Keough et al. (2016) used to determine copy numbers of the virion

within individual N. variabilis, however, its efficiency of this assay was not mentioned in the

study.

Transmission

The unique transmission properties of SVNV concur with the phylogenetic studies

distinguishing the virus from other well-characterized members of the genus. There are more

than 5000 thrips species described to date and only 17, belonging to the genera Frankliniella,

Thrips, Ceratothripoides, Scirtothrips, Dictyothrips, Neohydatothrips and Taeniothrips have

been confirmed as vectors of orthotospoviruses (Zhou and Tzanetakis, 2013; Riley et al., 2011;

Xu et al., 2017; Table 1). The primary vector of SVNV, Neohydatothrips variabilis (Beach) (Fig.

4) is the only vector species belonging to the subfamily Sericothripinae (Thysanoptera:

Thripidae); all the other genera belong to subfamily Thripinae (Thysanoptera: Thripidae). N.

variabilis is a common pest for soybean and cotton in the U.S. (Irizarry, 2016) and the

phylogenetic placement of N. variabilis mirrors the phylogenetic space of SVNV as an

orthotopovirus and may reflect the co-evolution of orthotopoviruses with their vectors (Ciuffo et

al., 2010). Recent studies have reported two other common thrips species Frankliniella tritici

(Fitch) (eastern flower thrips) and Frankliniella fusca (Hinds) (tobacco thrips) as vectors of

SVNV. Their transmission efficiencies, however, are much lower to that of N. variabilis suggesting that the latter has coevolved with the virus and acts as its primary vector in the field

(Keough et al., 2016). Another important orthotospovirus vector - Frankliniella occidentalis

(western flower thrips) is unable to transmit SVNV (Zhou, 2018). According to Keough et al.

(2016), SVNV-infected N. variabilis prefer to feed on non-infected leaflets and viruliferous females produce more offspring compared with their non-viruliferous counterparts. These

attributes may have contributed the rapid spread of SVNV in a short time span. Similar to other

orthotospoviruses, SVNV is considered to be transmitted in a propagative and persistent manner,

and the acquisition of the virus by its vectors is a life-stage-dependent process (Whitfield et al.,

2005).

Seed transmission has always been a major concern for virus diseases, especially for

seed-propagated crops like soybean, as it can act as the major route for long-distance dissemination of viruses (Hajimorad et al., 2015; Mink, 1993; Johansen et al., 1994; Hull, 2014).

Investigations on whether SVNV is a seed-transmissible virus has been conducted by different researchers in recent years with contradictory results. Groves and co-workers (Groves et al.,

2016) reported a 6% seed-transmission rate. In this study, a random seed sample obtained from a

seed lot of a commercial soybean variety was planted under controlled conditions. Leaf samples

collected from their seedlings were tested positive for SVNV using RT-PCR, but not ELISA.

Additional testing using arbitrarily selected plants from initial testing and repeated experiments

confirmed the presence of SVNV genome segments using RT-PCR and RNA-seq analysis. The authors therefore concluded that this is due to an asymptomatic, seed-transmissible SVNV isolate that is transmitted by soybean seeds at high rate. A study performed by Hajimorad et al.

(Hajimorad et al., 2015) on two soybean cultivars using over 2000 seeds derived from 20 SVNV- infected individual mother plants failed to detect the presence of SVNV using ELISA. They analyzed the genetic variation among SVNV isolate from infected mother plants and found the existence of a distinct isolate which has a unique amino acid mutation and branches separately from all other isolates indicating a relatively diverse virus population in the study. Considering the non-systemic movement of SVNV on soybean, the self-pollinating feature of soybean and the fact that SVNV infection occurs in the late growth stage, Hajimorad et al. concluded that it is very unlikely SVNV is transmitted by seed. The results of Hajimorad et al. are in agreement with our studies in which SVNV was not detected in over 600 seedlings germinated from seed collected from SVNV-infected mother plants of different cultivars growing in the field (Zhou, unpublished data).

Disease Management

Management of orthotospovirus-caused diseases has always presented a major challenge

(Bag et al., 2015; Pappu et al., 2009; Oliver and Whitfield, 2016). There is limited knowledge on many aspects of the biology and epidemiology of SVNV which are crucial for developing effective strategies for virus control and disease management. Since the primary and secondary vector species were well documented and more data became available for potential alternative hosts of the virus (Zhou and Tzanetakis, 2013; Escalante et al., 2018; Sikora et al., 2018; Irizarry et al., 2018; Keough et al., 2016), the current management options for soybean vein necrosis have focused on reducing the impact of thrips on soybeans and seeking potential virus reservoirs

in the field. Several studies have been conducted to determine the composition of thrips species

and population dynamics of SVNV vectors, especially for N. variabilis over seasons in different

geographic areas. Data shows N. variabilis is the most abundant vector species in Northern,

Midwestern and Southern U.S. states (Chitturi et al., 2018; Keough et al., 2018; Bloomingdale et

al., 2017). The fact that distinct seasonal trends of thrips migration were not detected based on

location in northern states suggests that virus vectors may not migrate from areas outside the

region, instead, they may colonize other host plants, especially perennial species to overwinter

during the absence of soybean and early in the growing season before moving to soybean

(Keough et al., 2018; Bloomingdale et al., 2017) which highlights the importance of finding and

eliminating local virus reservoirs. Considering the peak activity for the primary vector is either at

or prior to the occurrence of vein necrosis symptoms (Chitturi et al., 2018; Keough et al., 2018), it is possible to reduce disease incidence by managing the planting system or planting date, as suggested by Kleczewski (Kleczewski, 2018). No SVNV-resistant soybean cultivar has been identified at this moment although one study does show a differentiation in symptom intensities among cultivars (Anderson, 2017). Apart from the resistance directly to the pathogen, cultivars that have resistance to virus vector could also reduce disease incidence. Such resistance may result from physical or biochemical features of particular cultivars or the combination of the two factors. To search for potential thrips resistance, we screened soybean cultivars with differential levels of pubescence. Our study demonstrates the feeding damage caused by N. variabilis differs

among selected cultivars and is correlated to their pubescence levels (Zhou, 2018). The

effectiveness of chemical product on disease incidence including insecticide and seed treatment

has not be evaluated to date (Keough et al., 2018).

Summary

The continuous reports of SVNV during the past decade in major soybean producing

areas in North America draw attention from the scientific community. Research has been

conducted on different aspects of the virus and the disease it causes in order to better understand

its biology, epidemiology and estimate its impact on soybean yield. As a relatively new member

of the genus Orthotopovirus, SVNV represents a distinct evolutionary linage that has many

atypical molecular and biological characteristics. The inefficient movement of SVNV in soybean

and the homogeneous population structure across a wide geographic range indicate the virus is

most likely to be introduced from another host recently and have not adapted well to soybean, a

relatively new host of the virus (Zhou and Tzanetakis, 2013; Hajimorad et al., 2015; Khatabi et

al., 2012). On the other hand, the majority of alternative hosts characterized for SVNV to date

belongs to the Fabaceae family (Zhou, unpublished data), however, peanut (Arachis hypogaea) -

a host of several orthotopoviruses including GBNV, GRSV, GYSV and GCFSV is probably not

a host of SVNV (Zhou and Tzanetakis, 2013), although more cultivars need to be screened.

Collectively, these characteristics are in agreement with the orthotopovirus classification

proposed by Inoue and Sakurai (2007) which takes into consideration of host and vector

specificities, suggesting a host-related adaptation of the genus Orthotopovirus toward members of the Fabaceae family in the case of SVNV-BNeMV clade. Future investigations on the

function of viral proteins and host components may shed light on the special characteristics of

this virus-host-vector pathosystem and the evolutionary pathway of orthotopoviruses. A lack of a

reliable assay based on mechanical inoculation for SVNV infection of soybean is a bottleneck to

perform any study that requires a uniformed disease pressure, such as estimate of the response of

different soybean cultivars to virus infection and investigation on the synergistic interactions

between SVNV and other viruses prevalent in soybean. On the other hand, virus inoculation using viruliferous thrips as inoculum may be more effective in the identification of alternative hosts given that it could differentiate host preference of virus vectors. It was reported that SVNV infection on soybean reduces oil content of seeds but have minimal impact on the yield

(Anderson et al., 2017). The case of being able to detect the virus from seedlings derived from seeds collected from SVNV-positive mother plants has raised the profile of SVNV to a seed- transmissible virus (Groves et al., 2016). However, the presence of F. tritici – a SVNV virus vector in the greenhouse where the study took place and the lack of genetic information of the unique SVNV isolate that leads to asymptomatic and systemic infection mentioned does not allow for the further study of the mechanisms of virus seed invasion. Seed transmission is a complicated biological phenomenon which involves host genotype, physiological and developmental stage of the host, virus replication and movement as well as environmental conditions (Mink, 1993; Johansen et al., 1994; Hull, 2014; Sastry, 2013). Likewise, the impact of virus on yield and seed quality can also be affected by compounding factors including but not limited to cultivar genotype and timing of virus infection (Hopkins and Mueller, 1984; Ren et al.,

1997; Maestri et al., 1998; Filho et al., 2001; Byamukama et al., 2015). The fact that SVNV infection mostly occurs at the end of vegetative growth stages and the beginning of reproductive stages may mask its true impact on soybean. For those reason, additional studies are needed to investigate the effects of infection timing on disease symptoms intensity, yield, seed quality and potential of seed transmission.

0.1

Fig 1. Phylogenetic analysis based on alignment of all orthotopovirus RdRp amino acid sequences available in the National Center for Biotechnology Information (NCBI) Genbank as of September 2018. The dendrogram was produced in CLC Genomics Workbench 11.0.1 using the Neighbor-Joining algorithm with 1000 bootstrap replicates. Only bootstrap values greater than 90% are shown. The bar represents p-distance of 0.1. Virus acronyms used include: GBNV (groundnut bud necrosis virus; NP 619688), GRSV (groundnut ringspot virus; AST36116), INSV (impatiens necrotic spot virus; NP 619710), IYSV (iris yellow spot virus; YP 009241381), PolRSV (polygonum ringspot virus; AOO95317), TCSV (tomato chlorotic spot virus; YP 009408637), TSWV (tomato spotted wilt virus; NP 049362), WBNV (watermelon bud necrosis virus; YP 009505544), WSMoV (watermelon sliver mottle virus; AAW56420), ZLCV (zucchini lethal chlorosis virus; YP 009316178), BNeMV (bean necrotic mosaic virus; AEF56575), CCSV (calla lily chlorotic spot virus; YP 009449454), CSNV (chrysanthemum stem necrosis virus; AII20576), HCRV (hippeastrum chlorotic ringspot virus; CDJ79757), MeSMV (melon severe mosaic virus; YP 009346017); MYSV (melon yellow spot virus; YP 717933), PCSV (pepper chlorotic spot virus; YP 009345145); SVNV (soybean vein necrosis virus; ADX01591), TNSaV (tomato necrotic spot associated virus; AMY62790); TYRV (tomato yellow ring virus; AEX09314); TZSV (tomato zonate spot virus; YP 001740047), GCFSV (groundnut chlorotic fan-spot virus; AJT59689), MVBaV (mulberry vein banding associated virus; YP 009126736).

A B C

D E F

Fig 2. Disease symptoms caused by SVNV infection on soybean. A. Indistinguishable early symptom; B. Distinct early symptom; C. Uneven distribution of lesions on leaf blade; D. Mild disease symptom; E. Aggressive disease symptom; F. Scotched leaf blade due to virus infection.

A B C

Fig 3. Local lesions caused by SVNV infection on A. Cowpea; B. Medicago and C. Pigeon pea.

Table 1. Summary of thrips-orthotopovirus interactions confirmed by transmission studies (September 2018. Italics: assigned and; plain text: unassigned members of the genus Orthotospovirus. Citations in the parenthesis indicate references for additional thrips- orthotospoviruses interactions compared with Riley et al., 2011.).

Thrips Genera Thrips Species Orthotospovirus Vectored

Chrysanthemum stem necrosis virus Groundnut ringspot virus Impatiens necrotic spot virus Frankliniella Tomato chlorotic spot virus occidentalis Tomato spotted wilt virus Alstroemeria necrotic streak virus (Hassani-Mehraban et al., 2010)

Chrysanthemum stem necrosis virus Groundnut ringspot virus Frankliniella schultzei Groundnut bud necrosis virus Tomato chlorotic spot virus Frankliniella Tomato spotted wilt virus Groundnut ringspot virus Frankliniella intonsa Impatiens necrotic spot virus Tomato chlorotic spot virus Tomato spotted wilt virus Tomato spotted wilt virus Frankliniella fusca Impatiens necrotic spot virus Soybean vein necrosis virus (Keough et al., 2016)

Frankliniella gemina Tomato spotted wilt virus Groundnut ringspot virus

Frankliniella bispinosa Tomato spotted wilt virus Frankliniella zucchini Zucchini lethal chlorosis virus Frankliniella cephalica Tomato spotted wilt virus

Frankliniella tritici Soybean vein necrosis virus (Keough et al., 2016)

Calla lily chlorotic spot virus Groundnut bud necrosis virus Thrips palmi Melon yellow spot virus Watermelon silver mottle virus Thrips Tomato necrotic ringspot virus (Seepiban et al., 2011)

Iris yellow spot virus Thrips tabaci Tomato spotted wilt virus Tomato yellow fruit ring virus

Thrips setosus Tomato spotted wilt virus Groundnut bud necrosis virus Scirtothrips Scirtothrips dorsalis Groundnut chlorotic fan-spot virus Groundnut yellow spot virus

Table 1 (Cont.).

Thrips Genera Thrips Species Orthotospovirus Vectored

Ceratothripoides Ceratothripoides claratris Capsicum chlorosis virus Tomato necrotic ringsot virus (Seepiban et al., 2011)

Dictyothrips Dictyothrips betae Polygonum ringspot virus Neohydatothrips Neohydatothrips variabilis Soybean vein necrosis virus (Zhou and Tzanetakis, 2013; Keough et al., 2016)

Taeniothrips Taeniothrips. eucharii Hippeastrum chlorotic ringspot virus (Xu et al., 2017)

A B C

D E F

Fig. 4 Different life stages of Neohydatothrips variabilis (Beach). A. First instar larvae; B. Early second instar larvae; C. Late second instar larvae; D. Prepupa; E. Pupa and F. Adult.

Ali, A. and Abdalla, O. A. 2013. First report of Soybean vein necrosis virus in soybean fields of Oklahoma. Plant Disease. 97: 1664.

Anderson, N. R. 2017. Effect of soybean vein necrosis on soybean yield and seed quality, and symptom expression on soybean and alternative hosts. Thesis, Purdue University, West Lafayette, IN.

Anderson, N. R., Irizarry, M. D., Bloomingdale, C. A., Smith, D. L., Bradley, C. A., Delaney, D. P., Kleczewski, N. M., Sikora, E. J., Mueller, D. S., Wise, K. A. 2017. Effect of soybean vein necrosis on yield and seed quality of soybean. Canadian Journal of Plant Pathology. 39: 334-341.

Bag, S., Schwartz, H. F., Cramer, C. S., Havey, M. J. and Pappu, H. R. 2015. Iris yellow spot virus (Tospovirus: Bunyaviridae): from obscurity to research priority. Molecular Plant Pathology. 16: 224-237.

Bloomingdale, C., Irizarry, M. D., Groves, R. L., Mueller, D. S. and Smith, D. L. 2017. Seasonal population dynamics of thrips (Thysanoptera) in Wisconsin and Iowa soybean fields. Journal of Economic Entomology. 110: 133-141.

Bruenn, J. A. 2003. A structural and primary sequence comparions of the viral RNA-dependent RNA polymerases. Nucleic Acids Research. 31: 1821–1829.

Byamukama, E., Robertson, A. E. and Nutter, F. W. Jr. 2015. Bean pod mottle virus time of infection influences soybean yield, yield components, and quality. Plant Disease. 99: 1026–1032.

Caruthers, J. M. and McKay, D. B. 2002. Helicase structure and mechanism. Current Opinion of Structural Biology. 12: 123-133.

Chen, T. C., Li, J. T., Fan, Y. S., Yeh, Y. C., Yeh, S. D. and Kormelink, R. 2013. Molecular characterization of the full-length L and M RNAs of Tomato yellow ring virus, a member of the genus Tospovirus. Virus Genes. 46: 487-495.

Chitturi, A., Conner, K., Sikora, E. J. and Jacobson, A. L. 2018. Monitoring seasonal distribution of thrips vectors of soybean vein necrosis virus in Alabama soybeans. Journal of Economic Entomology. https://doi.org/10.1093/jee/toy237

Ciuffo, M., Mautino, G. C., Bosco, L., Turina, M. and Tavella, L. 2010. Identification of Dictyothrips betae as the vector of Polygonum ring spot virus. Annual of Applied Biology. 157: 299-307.

Conner, K., Sikora, E. J., Zhang, L., Burmester, C. 2013. First report of soybean vein necrosis- associated virus affecting soybeans in Alabama. Plant Health Progress. https://doi.org/10.1094/PHP-2013-0729-03-BR

Cortez, I., Aires, A., Pereira, A. M., Goldbach, R., Peters, D. and Kormelink, R. 2002. Genetic organization of Iris yellow spot virus M RNA: indications for functional homology between the G(C) glycoproteins of tospoviruses and animal-infecting bunyaviruses. Archive of Virology. 147: 2313–2325.

de Oliveira, A. S., Bertran, A. G. M., Inoue-Nagata, A. K., Nagata, T., Kitajima, E. W. and Oliveira Resende, R. 2011. An RNA-dependent RNA polymerase gene of a distinct Brazilian tospoviurs. Virus Genes. 43: 385-389.

de Oliveira, A. S., Melo, F. L., Inoue-Nagata, A. K., Nagata, T., Kitajima, E. W. and Resende, R. O. 2012. Characterization of bean necrotic mosaic virus: a member of a novel evolutionary lineage within the Genus Tospovirus. PLoS One 7: e38634.

Dunn, E. F, Pritlove, D. C., Jin, H. and Elliott, R. M. 1995. Transcription of a recombinant bunyavirus RNA template by transiently expressed bunyavirus proteins. Virology. 211: 133-143.

Escalante, C., Bollich, P. and Valverde, R. 2018. Soybean vein necrosis-associated virus naturally infecting yard-long bean (Vigna unguiculata ssp. Sesquipedalis) and soybean in Louisiana. Plant Disease. 102: 10.

Filho, M. M., Destro, D., Miranda, L. A., Spinosa, W. A., Carrao-Panizzi, M. C. and Montalvan, R. 2001. Relationships among oil content, protein content and seed size in soybeans. Brazilian Archive of Biology and Technology. 44: 23–32.

Flick, R. and Pettersson, R. F. 2001. Reverse genetics system for Uukuniemi virus (Bunyaviridae): RNA polymerase I-catalyzed expression of chimeric viral RNAs. Journal of Virology. 75: 1643-1655.

Flick, K., Hooper, J. W., Schmaljohn, C. S., Pettersson, R. F., Feldmann, H. and Flick, R. 2003. Rescue of Hantaan virus minigenomes. Virology. 306: 219-224.

Groves, C., German, T. L., Dasgupta, R., Mueller, D. and Smith, D. L. 2016. Seed transmission of soybean vein necrosis virus: the first tospovirus implicated in seed transmission. PLoS One. 11: e0147342.

Hajimorad, M. R., Halter, M. C., Wang, Y., Staton, M. E. and Hershman, D. E. 2015. Evaluation of seed transmissibility of Soybean vein necrosis-associated virus in two soybean cultivars grown under field conditions. Journal of Plant Pathology and Microbiology. 6: 278-283.

Han, J., Domier, L. L., Dorrance, A. E. and Qu, F. 2013. First report of Soybean vein necrosis- associated virus in Ohio soybean fields. Plant Disease. 97: 693.

Hassani-Mehraban, A., Botermans, M., Verhoeven, J. T., Meekes, E., Saaijer, J., Peters, D., Goldbach, R. and Kormelink, R. 2010. A distinct tospovirus causing necrotic streak on Alstroemeria sp. In Colombia. Archive of Virology. 155: 423-428.

Hopkins, J. D. and Mueller, J. 1984. Effect of bean pod mottle virus on soybean yield. Journal of Economic Entomology. 77: 943–947.

Huang, K. S., Tai, C. H., Cheng, Y. H., Lin, S. H., Chen, T. C. and Jan, F. J. 2017. Complete nucleotide sequences of M and L RNAs from a new pepper-infecting tosppovirus, pepper chlorotic spot virus. Archive of Virology. 162: 2109-2113.

Hull R. (2014) Plant Virology. Fifth edition. Academic Press, New York.

Inoue, T. and Sakurai, T. 2007. The phylogeny of thrips (Thysanoptera: Thripidae) based on partial sequences of cytochrome oxidase I, 28S ribosomal DNA and elongation factor-1 α and the association with vector competence of tospoviruses. Apploed Entomology and Zoology. 42: 71-81.

Jacobs, J. L. and Chilvers, M. I. 2013. First report of Soybean vein necrosis virus on soybeans in Michigan. Plant Disease. 97: 1387.

Johansen, I. E., Edwards, M. C., Hampton, R. O. 1994. Seed transmission of viruses – current perspectives. Annual Review pf Phytopathology. 32: 363-386.

Kainz, M., Hilson, P., Sweeney, L., DeRose, E. and German, T. L. 2004. Interaction between Tomato spotted wilt virus N protein monomers involves nonelectrostatic forces governed by multiple distinct regions in the primary structure. Phytopathology. 94: 759–765.

Keough, S., Han, J., Shuman, T., Wise, K. and Nachappa, P. 2016. Effects of Soybean Vein Necrosis Virus on Life History and Host Preference of Its Vector, Neohydatothrips variabilis, and Evaluation of Vector Status of Frankliniella tritici and Frankliniella fusca. Journal of Economic Entomology. 109: 1979-1987.

Keough, S., Danielson, J., Marshall, J. M., Lagos-Kutz, D., Voegtlin, D. J., Srinivasan, R. and Nachappa, P. 2018. Factors affecting population dynamics of thrips vectors of soybean vein necrosis virus. Environmental Entomology. 47: 734-740.

Khatabi, B., Wen, R. H., Hershman, D. E., Kennedy, B. S., Newman, M. A. and Hajimorad. M. R. 2012. Generation of polyclonal antibodies and serological analyses of nucleocapsid protein of soybean vein necrosis associated virus: a distinct soybean infecting tospovirus serotype. European Journal of Plant Pathology. 133: 783-790.

Kleczewski, N. 2016. Research updates on soybean vein necrosis virus. http://extension.udel.edu/fieldcropdisease/2016/01/20/research-updates-on-soybean-vein- necrosis-virus.

Kleczewski, N. 2018. Prevalence and cropping systems impacts on soybean vein necrosis disease in Delaware soybeans. Plant Health Progress. 19: 11-12.

Kormelink, R., De Haan, P., Meurs, C., Peters, D. and Goldbach, R. 1992. The nucleotide sequence of the M RNA segment of tomato spotted wilt virus, a bunyavirus with two ambisense RNA segments. Journal of General Virology. 73: 2795–2804.

Kukkonen, S. K. J., Vaheri, A. and Plyusnin, A. 2005. L protein, the RNA-dependent RNA polymerase of hantaviruses. Archive of Virology. 150: 533–556.

Lokesh, B., Rashmi, P. R., Amruta, B. S., Srisathiyanarayanan, D., Murthy, M. R. N. and Savithri, H. S. 2010. NSs encoded by groundnut bud necrosis virus is a bifunctional enzyme. PLoS One 5: e9757.

Irizarry, M. 2016. Soybean vein necrosis virus: impacts of infection on yield loss and seed quality and expansion of plant host range. Thesis, Iowa State University, Ames, IA.

Irizarry, M. D., Elmore, M. G., Batzer, J. C., Whitham, S. A., Mueller, D. S. 2018. Alternative hosts for soybean vein necrosis virus and feeding perferences of its vector soybean thrips. Plant Health Progress. 19: 176-181.

Maestri, D. M., Guzman, G. A., Giorda, L. M., Labuckas, D. O. 1998. Correlation between seed size, protein and oil contents, and fatty acid composition in soybean genotypes. Grasas y Aceites. 49: 450–453.

Melcher, U. 2000. The “30 K” superfamily of viral movement proteins. Journal of General Virology. 80: 257–266.

Mink, G. I. 1993. Pollen and seed-transmitted viruses and viroids. Annual Review pf Phytopathology. 31: 375-402.

Mushegian, A. R. and Koonin, E. V. 1993. Cell-to-cell movement of plant viruses. Archive of Virology. 133: 239–257.

Oliver, J. E. and Whitfield, A. E. 2016. The genus tospovirs: emerging bunyaviruses the threaten food security. Annual Review of Virology. 29: 101-124.

Pappu, H. R., Jones, R. A. and Jain, R. K. 2009. Global status of tospovirus epidemics in diverse cropping systems: successes achieved and challenges ahead. Virus Research. 141: 219-236.

Ren, Q., Pfeiffer, T. W. and Ghabrial, S. A. 1997. Soybean mosaic virus incidence level and infection time: Interaction effects on soybean. Crop Science. 37: 1706–1711.

Riley, D. G., Joseph, S. V., Srinivasan, R. and Stanley, D. 2011. Thrips vectors of tospoviruses. Journal of Integrated Pest Management. 1: 1-10.

Roberts, A., Rossier, C., Kolakofsky, D., Nathanson, N. and Franscisco, G. S. 1995. Completion of the La crosse virus genome sequence and genetic comparisons of the L proteins of the Bunyaviridae. Virology. 206: 742–745.

Sastry, K. S. 2013. Seed-borne plant virus disease. Springer, New York.

Seepiban, C., Gajanandana, O., Attathom, T. and Attathom, S. 2011. Tomato necrotic ringspot virus, a new tospovirus isolated in Thailand. Archive of Virology. 156: 263-274.

Sherwood, J. L., German, T. L., Moyer, J. W., Ullman, D. E. and Whitfield, A. E. 2000. Tomato spotted wilt. In: Maloy OC, Murray TD (ed) Encyclopedia of plant pathology. Wiley, New York, pp 1034–1040.

Sikora, E. J., Conner, K. N. and Jacobson, A. L. 2018. Incidence of soybean vein necrosis virus in Alabama soybean field. Plant Health Progress. 19: 76-81.

Silva, M. S., Maertins, C. R. F., Bezerra, I. C., Nagata, T., De Ávila, A. C. and Resende, R. O. 2001. Sequence diversity of NSm movement of protein of tospoviruses. Archive of Virology. 146: 1267–1281.

Smith, D. L., Fritz, C., Watson, Q., Willis, D. K., German, T. L., Phibbs, A., Mueller, D., Dittman, J. D., Saalau-Rojas, E. and Whitham, S. A. 2013. First report of Soybean vein necrosis disease caused by soybean vein necrosis-associated virus in Wisconsin and Iowa. Plant Disease. 97: 693.

Takeda, A., Sugiyama, K., Nagano, H., Mori, M., Kaido, M., Mise, K., Tsuda, S. and Okumo, T. 2002. Identification of a novel RNA silencing suppressor, NSs protein of Tomato spotted wilt virus. FEBS Letters. 532: 75–79.

Tzanetakis, I. E., Wen, R. H., Newman, M. 2009. Soybean vein necrosis virus: A new threat to soybean production in Southeastern United States? Phytopathology. 99: 131.

Whitfield, A. E., Ullman, D. E. and German, T. L. 2005. Tospovirus-thrips interactions. Annual Review of Phytopathology. 43: 459-489.

Xu, Y., Gao, X., Jia, Z., Li, W., Hu, J., Li, Y., Li, Y. and Liu, Y. 2017. Identification of Taeniothrips eucharii (Thysanoptera: Thripidae) as a Vector of Hippeastrum chlorotic ringspot virus in Southern China. Plant Disease. 101: 1597-1600.

Zhou, J., Kantartzi, S. K., Wen, R. H., Newman, M., Hajimorad, M. R., Rupe, J. C., Tzanetakis, I. E. 2011. Molecular characterization of a new tospovirus infecting soybean. Virus Genes. 43: 289-295.

Zhou, J. 2012. Characterization and epidemiology of Soybean vein necrosis associated virus. Thesis, University of Arkansas, Fayettille, AR.

Zhou, J. and Tzanetakis, I. E. 2013. Epidemiology of Soybean vein necrosis-associated virus. Phytopathology. 103: 966-971.

Zhou, J., Aboughanem-Sabanadzovic, N., Sabanadzovic, S. and Tzanetakis, I. E. 2018. First report of soybean vein necrosis virus infecting kudzu (Pueraria montana) in the United States of America. Plant Disease. 102: 1674.

Zhou, J. 2018. Epidemiological Studies of soybean vein necrosis virus and potential resistance mechanisms to its vector Neohydatothrips variabilis (Beach). PhD Dissertation, University of Arkansas, Fayetteville, AR. Submitted.

Chapter II

Identification of Alternative Hosts in Arkansas and Evaluation of Frankliniella occidentalis as a Potential Vector of Soybean Vein Necrosis Virus

Abstract

Alternative hosts play an important role in disease epidemiology. They enable pathogens,

including the focus of this study soybean vein necrosis virus (SVNV), to accumulate in the

environment and spread to broader geographic areas, persist for extended periods of time and act

as green-bridges in the absence of the primary host. In order to better understand the role of weeds in the virus disease cycle, surveys were conducted to determine the presence of SVNV among 32 weed species collected from soybean fields in Arkansas. The presence of the virus in

individual plants was screened using dot blot immunoassay and/or DAS-ELISA and confirmed by RT-PCR. To verify the data obtained from the field survey, SVNV was transmitted back to plants using viruliferous Neohydatothrips variabilis in greenhouse experiments and confirmed that kudzu (Pueraria montana), present in millions of acres in Southeastern United States, can sustain SVNV replication in a systemic manner. Frankliniella occidentalis, the most economically important vector of orthotospoviruses, was evaluated as potential vector of SVNV and was unable to transmit the virus. Our findings suggest that kudzu is likely to play a role in the epidemiology of the virus and the disease.

Introduction

Soybean is one of the most important row crops worldwide and has been successfully

grown in North America since the early 1920s. In the U.S., soybean is planted in 38 states with

an annual production exceeding 90 million metric tons (Hartman, 2015). The environmental

conditions that benefit soybean also favor diseases and pests. New virus-like symptoms were

first observed in Arkansas and Tennessee in 2008 and were subsequently reported widely spread

in other states including Illinois, Wisconsin, Michigan, Ohio, Pennsylvania, Delaware, Kansas,

Oklahoma, Kentucky, Missouri, Mississippi, Louisiana and Alabama (Tzanetakis et al., 2009;

Zhou, 2012; Ali and Abdalla, 2013; Conner et al., 2013; Han et al., 2013; Jacobs and Chilvers,

2013; Smith et al., 2013; Kleczewski, 2016; Escalante et al., 2018). The disease was named

soybean vein necrosis (SVN) as symptoms initiate with vein clearing along the main veins,

which become chlorotic as leaves expand and eventually become necrotic. Double-stranded

RNA was extracted from symptomatic soybean leaves and a new orthotospovirus, designated as soybean vein necrosis virus (SVNV) was verified as the causal agent (Zhou et al., 2011, Zhou and Tzanetakis, 2013). The virus has been confirmed in at least 22 states across the U.S. as well as Canada and Egypt (Abd El-Wahab and El-Shazly, 2017) and vein necrosis (SVN) has become the most prevalent virus disease in North America (Zhou and Tzanetakis, 2013).

The widespread occurrence of SVN in the major soybean producing areas in North

America makes it essential to study the epidemiology of the virus in depth. Studies performed during 2009 to 2012 identified Ipomoea hederacea Jacq. (ivyleaf morning glory), an indigenous weed in soybean fields as a natural host and Neohydatothrips variabilis (Beach) (soybean thrips) as a natural vector of the virus, respectively (Zhou and Tzanetakis, 2013). This provides basic epidemiological information, however, is not sufficient to explain how SVNV could cause an

epidemic in such a short period of time given the crop-free period in North America. Previous studies revealed that weeds function as virus sources and shelter of their vectors, providing a green bridge in the absence of their primary hosts (Hsu et al., 2011; Smith et al., 2011).

Likewise, indigenous weeds in soybean fields may also play an important role in the dissemination of SVNV.

Field surveys were conducted to determine the incidence of the virus in indigenous weeds collected from soybean fields in Arkansas. This information was used to examine whether N. variabilis survive and multiply on host weed species, and whether those hosts allow for the systemic movement of the virus. In addition, we tested the hypothesis that Frankliniella occidentalis can transmit the virus.

Materials and Methods

Field survey

Soybean fields located at the Arkansas Agricultural Research and Extension Center

(Fayetteville, U. S.) were selected for sample collection due to the high SVNV incidence between 2009-2012. Thirty-one common weed species including eight monocotyledon and twenty-three dicotyledon species were collected between June-October 2013. Leaf tissue from at least six areas including upper, middle and lower canopy were collected and pooled together as one sample so as to detect the virus in non-systemic hosts. In another field survey, kudzu

(Pueraria montana) was collected from different countries of Mississippi during September

2015 and tested for the presence of SVNV.

Greenhouse study

Neohydatothrips variabilis colony and virus culture

Adults N. variabilis were collected using mini-aspirator from upper canopy of soybean plants in the field, released to soybean seedlings grown in 20” by 10” planting trays and kept in growth chamber under controlled environment (27°C, Light: Dark = 16h: 8h). Trays were replaced with new seedlings every three weeks. Adults and larvae from old plants were pooled and placed onto new ones for thrips rearing. SVNV was maintained within growth chambers as viruliferous thrips propagated and plants were tested periodically using dot blot immunoassay and RT-PCR to confirm the presence of the virus.

Virus inoculation

N. variabilis adults (150-200) were pooled on SVNV-tested detached soybean trifoliates for rearing larvae under controlled environment (22°C, Light: Dark = 16h: 8h). Larvae hatched from the leaf blade within 24 h were transferred individually to SVNV-infected tissue using a size 0 camel painting brush and allowed to feed for 48 h for virus acquisition. A cohort of viruliferous larvae (15-20) were transferred on seedlings geminated from seeds harvested from weed species that were tested positive for SVNV in field survey. The seeds were collected from the same fields as positive samples to minimize potential biotype variability that could affect

SVNV replication.

Detection

Field samples collected from the field were tested with dot blot immunoassay for the incidence of SVNV infection. Briefly, 0.2 g pooled plant tissue was ground in 2 ml 1×PBS buffer and centrifuged to precipitate plant debris. Samples were dotted on GE Amersham

Protran NC nitrocellulose membrane (Thermo Fisher Scientific, Waltham, MA). The dotted

membranes were dried and washed with PBS buffer then transferred into blocking buffer (PBS,

0.1% Tween-20 and 5% non-fat milk) for 1 h followed by washing with PBS-Tween (PBS buffer containing 0.1% Tween-20) and incubation in antibody solution (Agdia Inc, IN, U.S.;

1:1000 dilution in PBS) for 1 h. The membrane was then transferred to goat anti-rabbit alkaline phosphatase conjugate solution (Sigma, MO, USA; 1:5000 dilution in PBS containing 0.2% non-fat milk) and incubated for 1 h. After wash with PBS-Tween, the membrane was transferred to substrate buffer ( 0.1 M Tris, pH 9.5, 0.1M NaCl and 5 mM MgCl2, containing 0.67% NBT and 0.33% BCIP). The reaction was stopped by transferring membrane to deionized water.

Double antibody sandwich ELISA (DAS-ELISA) was performed as described by Clark et al. (1977) to test SVNV incidence on kudzu. Briefly, SVNV antibody stock (Agdia Inc., IN,

U.S.) was diluted 1:1000 with coating buffer according to manufacturer’s instruction. One hundred microliter of diluted antibody was added to each well of polystyrene plate and incubated at room temperature for 3 – 5 h. The plate was then emptied and tapped dry before adding 200 µl blocking buffer, which was then incubated at room temperature for 1 h. During the incubation, 100 mg SVNV infected leaf tissue was ground thoroughly in 2 ml sample buffer for further use. After tapping dry the plate, 100 µl plant sap was added in each well and left overnight at 4 °C. The plate was sealed within air-tight bag to prevent evaporation. Plate was emptied and washed three times with wash buffer. After the final wash, conjugate was added to each well and incubated at room temperature for at least 4 h. The plate was washed thoroughly as above to ensure that all unbound conjugate was removed to minimize background reactions.

One tablet of p-nitrophenyl phosphate (5 mg; Sigma-Aldrich, MO, U.S.) was completely dissolved in 10 ml substrate buffer and 100 µl were added to each well. Incubation was done at room temperature until color development.

Positive samples from dot blot assay and/or ELISA were verified by RT-PCR. Briefly, total nucleic acids were extracted from pooled plant tissue as mentioned and used as template for cDNA synthesis (Zhou and Tzanetakis, 2013). RNase-free DNaseI was used to degrade DNA in purified total nucleic acids according the manufacturer’s instruction (Fermentas, MD, U.S.). The amount and quality of RNA were evaluated with a spectrophotometer (Nanodrop, Thermo

Fisher Scientific, DE, U.S.). Samples with good quality (260/280 >1.8) were used for downstream reactions. Reverse transcription was performed in a 25 μl reaction consisting of template RNA, 1 μl dNTPs (10 mM), 0.5 μl Ldet F/Sdet R (20 μM for each; Table 1), 5 μl

5×reverse transcriptase buffer (250 mM Tris-HCl, pH 8.3, 375 mM KCl, 15 mM MgCl2, 50 mM

DTT), 50 U reverse transcriptase (200 U/μl, Maxima, Fermentas) and 6 U RNase inhibitor (40

U/μl, RiboLock, Fermentas). The reaction was carried out in a thermocycler (Thermo C-1000,

Thermo Fisher Scientific) with incubation at 50°C for 60 min followed by enzyme deactivation at 80°C for 5 min. The synthesized cDNA was used as template for detection PCR. Detection primers SVNV-NP F and SVNV-NP R (Table 1) were used in a 25 μl PCR consisting of 0.5 μl dNTPs (10 mM), 0.5 ul reverse and forward primers (20 uM for each), 0.2 U Taq polymerase (5

U/ul; Genscript, NJ, U.S.) and 2.5 ul cDNA template. The reaction was initiated by denaturation at 94°C for 2min, followed by 40 cycles of denaturation at 94°C for 30 s, annealing at 55°C for

10 s, extension at 72°C for 30 s, and a final extension at 72°C for 10 min. Amplification products were visualized in 2% agrose gel in 0.5×TBE (40 mM Tris-HCl, 45 mM boric acid, 1 mM EDTA, pH7.2) post-stained with GelRed®(Biotium, CA, U.S.). Amplicons with corresponding size were sequenced (Eton Bioscience Inc. NC, U.S.) and aligned with the reference isolate (GenBank Acc. No. GU722319). Weeds exposed to viruliferous thrips feeding

were tested two weeks post transmission using both dot blot immunoassay and RT-PCR. Plants

tested negative for the virus were tested again a week later.

Alternative virus vector

Colony of Frankliniella occidentalis (western flower thrips) was kindly provided by Dr.

Anna Whitfield from North Caroline State University under a USDA permit and maintained on

green bean pods as instructed (Whitfield et al., 2008) with minor modifications. Briefly, larvae

hatched from the original colony were transferred to fresh bean pods and placed in a 12 oz deli

cup with its lid being partially cut off and replaced with a piece of thrips-proof fine mesh for

ventilation and prevention of the escape of thrips. Bean pods were replaced every other day till

adults emerged and laid eggs. Larvae with the age of 0-24 h old were collected and fed on

SVNV-infected tissue for 48 h to allow virus acquisition. A cohort of potentially viruliferous

larvae (20) was used as virus inoculum to inoculate the leaflets of unifoliate ‘Hutcheson’

seedling. The presence of SVNV was detected using RT-PCR as mentioned above.

Results

Field survey

The survey lasted for five months to include early-, middle- and late season species in the field. For each weed species, at least 30 individuals were collected and tested for virus infection

(Table 2). Three kudzu samples and one broadleaf signalgrass were tested positive for SVNV using ELISA and dot blot, respectively. In order to verify these results, virus specific RT-PCR tests were performed using nucleic acids extracted from all positive samples along with SVNV- positive and negative controls. A DNA band of 236 bp corresponding to NSs gene of SVNV was amplified from three ELISA-positive kudzu samples and positive control, but not from neither

dot blot-positive broadleaf signalgrass nor negative control. The PCR products were directly sequenced in both directions and confirmed to be SVNV-specific as they shared 96-99% nt identity with corresponding region of the SVNV reference isolate (GenBank Acc. No.

GU722319).

Greenhouse study

First instar larvae of N. variabilis were fed on symptomatic soybean for virus acquisition and transferred to eight SVNV-free kudzu seedlings as described (Zhou and Tzanetakis, 2013).

Leaf and root tissues were randomly collected from virus – inoculated and mock-inoculated

kudzu (Fig. 1), respectively, at six weeks post inoculation. Total nucleic acids were isolated and

RT-PCR was performed using primer set SdetF/ SdetR (Zhou et al., 2011). Amplification

products corresponding to the complete NP gene (834 bp) were obtained in both leaf and root

tissues from five inoculated plants and shared 97-99% identity with the reference isolate

(GenBank Acc. No. GU722319), whereas there was no amplification from mock-inoculated controls. Those results were further confirmed by dot blot immunoassay with infected plants remaining asymptomatic for at least two months post inoculation.

Alternative virus vector

First instar larvae of F. occidentalis were fed on SVNV-infected soybean leaves for virus acquisition followed by inoculation on 20 virus-tested ‘Hutcheson’ seedlings. Larvae fed on

SVNV-free plants were used as negative control. SVNV symptoms were not observed on either inoculated or systemic leaves and the virus was not detected at 4 or 6 weeks post inoculation in any of the plants using RT-PCR as mentioned above.

Discussion

SVN has been reported in major soybean producing areas in the U.S. and Canada with an increasing number of states joining the list in recent years (Tzanetakis et al., 2009; Zhou, 2012;

Ali and Abdalla, 2013; Conner et al., 2013; Han et al., 2013; Jacobs and Chilvers, 2013; Smith et al., 2013; Kleczewski, 2016; Escalante et al., 2018). However, little is known about the underlying factors driving the disease epidemics. In this study, we conducted field surveys on common weed species to determine whether they can sustain SVNV replication. Field data was used to design greenhouse experiments to verify results. Kudzu (P. montana) was identified as a new host of the virus. Frankliniella occidentalis was evaluated and eliminated as an alternative vector of the virus.

Among all weed species included in the field survey, broadleaf signal grass and kudzu were tested positive for SVNV using dot blot immunoassay and ELISA, respectively. Signal grass was proved negative for virus infection when tested with RT-PCR; this result was further confirmed by the greenhouse study where viruliferous thrips were used as the source of inoculation. The false positive sample could be due to the non-specific binding between SVNV antibodies and signal grass proteins or contamination with viruliferous soybean thrips. On the other hand, kudzu was tested positive in both field and greenhouse studies verifying it is an alternative host of SVNV. To evaluate the potential role of kudzu in disease epidemics, SVNV distribution in the new host was examined. Viruliferous N. variabilis fed on seedlings propagated as plants grow, leading to dense and sporadically distributed feeding scars on the leaf surface.

The feeding damage was at such a high level that it was not possible to obtain systemic tissue that was not damaged by thrips (Fig 1). We therefore collected leaf and root tissue from thrips- infested plants to determine the distribution of SVNV through the entire plant. The presence of

the virus was confirmed in both above and under the soil line in several plants making kudzu an

asymptomatic and systemic host of SVNV.

A growing body of evidence support the idea of evolved avirulence: in most cases, the

fittest parasites evolve toward avirulence among members of reservoir resulting in long – term

evolutionary relationship (Childs and Peters, 1993; Garnet and Antia, 1994; Yates et al., 2002).

Such coevolution results is a trade-off between host death and probability of virus transmission,

and viruses are maintained typically asymptomatic infections in such hosts (Garnet and Antia,

1994; Palmieri, 1982; Peterson et al., 2004). This line of reasoning suggests that compared with

symptomatic alternative hosts, plant species that sustain SVNV replication and remain

asymptomatic serves as a favorable source for the perpetuation of viruses in nature. Up to date,

twelve (12) species have been reported as alternative hosts of the virus; infection on ivy leaf

morning glory and yard-long bean were confirmed in field samples whereas the rest were

experimental hosts (Escalante et al, 2018; Irizarry et al., 2018; Zhou and Tzanetakis, 2013; Zhou

et al., 2018). The extension of SVNV host range to include an asymptomatic leguminous species

that favors rearing of virus vector provides new perspectives on SVNV epidemiology. The fact

that the virus moves systemically in kudzu and remains asymptomatic suggests kudzu could

function as a preferred host in nature. On the other hand, the recurring and expanding geographic

range of the virus point to green bridges that serve as external and within-crop sources for the

virus and/or viruliferous thrips. Irizarry et al. (2018) pointed out weed species such as alfalfa,

crimson clover and red clover as preferred hosts for N. variabilis. However, our studies failed to detect the presence of SVNV on these species; questioning their role in disease epidemics. A thorough understanding of Fabaceae species in the dissemination of SVNV requires field surveys across a wider geographic range to include diverse plant genotypes. On the other hand, kudzu, a

perennial weed present in millions of acres in Southeastern U.S. with overlapping geographic range to major soybean production areas could act as a reservoir of SVNV. It could provide overwintering or early season population growth habitats for viruliferous thrips prior to moving to soybean fields.

Similar to the alternative hosts, alternative vectors may have significant impact on disease epidemics. Other than the primary vector of SVNV, Keough et al. (2016) identified

Frankliniella tritici and Frankliniella fusca as vectors of the virus, yet both species showed low transmission efficiencies compared to N. variabilis. In the thrips - tospovirus interactions, it is not uncommon that one virus species is transmitted by multiple thrips species, or one thrips species is capable of transmitting more than one virus species (Amin et al., 1981; Ohnishi et al.,

2006; Premachandra et al., 2005; Riley et al., 2011). For example, Tomato spotted wilt virus is transmitted by thrips species including F. occidentalis, Thrips tabaci, Frankliniella schultzei, F. fusca, Frankliniella intonsa, Frankliniella bispinosa, Thrips setosus, Frankliniella gemina and

Frankliniella cephalica. On the other hand, F. occidentalis and F. schultzei are the vectors of five tospoviruses, respectively (Riley et al., 2011). This multi – virus/multi – vector association not only facilitates the spread of viruses to a wider host range but also have a major impact on virus evolution given that it increases the likelihood of reassortment of virus segments; leading to the suppression or breakdown of resistance (Qiu, W, and Moyer, J. W. 1999; Tentchev et al.,

2011; Margaria., 2015). In this study, we provided evidence that F. occidentalis is unable to transmit SVNV. As polyphagous thrips species, F. occidentalis feed and reproduce on at least 60 plant families (Tommasini and Maini, 1995; Loomans et al., 2006). By eliminating this species from the list of SVNV alternative vector, we could focus on the other virus vectors for virus control and disease management. From the aspect of evolution of tospoviruses, this finding also

minimizes the possibility of the genetic reassortment between SVNV and other tospoviruses that

are transmitted by F. occidentalis.

Table 1. List of oligonucleotide primers and probe used in RT-PCR. F-forward primer, R- reverse primer.

Primer Name Nucleotide Sequence (5’-3’)

Ldet F GAGCCCATAAACCTGTCTGC

Sdet R GATTAAACAGAAAACTCCTTTG

SVNV-NP F ACTTGTGCAAGCTTATGGT

SVNV-NP R GAAATGATTCCAATCTGTTC

Table 2. Weed species that were tested positive for soybean vein necrosis virus using dot blot immunoassay and/or ELISA.

Common Name Scientific Name Family Positive/ Number Tested Palmer amaranth Amaranthus palmeri Amaranthaceae 0/38 Barnyardgrass Echinochloa crus-galli Poacea 0/42 Horseweed Erigeron Canadensis Astera ceae 0/37 Prickly sida Sida spinose Malvaceae 0/32 Hemp sesbania Sesbania herbacea Fabaceae 0/30 Italian ryegrass Lolium multiflorum Poacea 0/49 Yellow nutsedge Cyperus esculentus Cypera ceae 0/60 Sicklepod Senna obtusifolia Fabaceae 0/44 Giant ragweed Ambrosia trifida Asteraceae 0/34 Broadleaf signalgrass Urochloa platyphylla Poacea 1/105 Common waterhemp Amaranthus tuberculatus Amaran thaceae 0/30 Hairy crabgrass Digitaria sanguinalis Poacea 0/43 Henbit Lamium amplexicaule Lamiace ae 0/36 Horsenettle Solanum carolinense Solanaceae 0/38 Spreading dayflower Commelina diffusa Commelinaceae 0/35 Swamp smartweed Persicaria hydropiperoides Polygonaceae 0/36 Hophornbeam Acalypha ostryifolia Euphorbiaceae 0/32 Copperleaf Cutleaf evening Oenothera laciniata Onagraceae 0/41 Primrose Spotted spurge Euphorbia maculate Euphorbiaceae 0/32 Common ragweed Ambrosia artemisiifolia Asteraceae 0/62 Eclipta Eclipta prostrata Asteraceae 0/43 Curly dock Rumex crispus Polygonaceae 0/30 Chickweed Stellaria media Caryophyllaceae 0/31 Common purslane Portulaca oleracea Portulacaceae 0/30 Shepherd's purse Capsella bursa-pastoris Brassicaceae 0/33 Spurred anoda Anoda cristata Malvaceae 0/45 Alfafa Medicago sativa Fabaceae 0/94 Johnsongrass Sorghum halepense Poaceae 0/65 White clover Trifolium repens Fabaceae 0/316 Red clover Trifolium pretense Fabaceae 0/65 Wild onion Allium spp. Amaryllidaceae 0/50 Kudzu Pueraria montana Fabaceae 3/85

Fig 1. Kuzdu, Pueraria montana inoculated with viruliferous N. variabilis at four (left) and six (right) weeks post inoculation.

References Abd El-Wahab, A.S. and El-Shazly, M. A. 2017. Identification and characterization of soybean vein necrosis virus (SVNV): a newly isolated thrips-borne tospovirus in Egypt. Journal of Virology Sciences. 1: 76-90. Ali, A. and Abdalla, O. A. 2013. First report of Soybean vein necrosis virus in soybean fields of Oklahoma. Plant Disease. 97: 1664. Amin, P. W., Reddy, D. V. R. and Ghanekar. 1981. Transmission of Tomato spotted wilt virus, the causal agent of bud necrosis of peanut, by Scirtothrips doralis and Frankliniella schultzei. Plant Disease. 65: 663-665. Clark, M. F. and Adams, A. N. 1977. Characteristics of the microplate method of enzyme-linked immunosorbent assay for the detection of plant viruses. Journal of General Virology. 34: 475- 483. Chandler, L., Hogge, G., Endres, M., Jacoby, D. R., Nathanson, N. and Beaty, B. J. 1991. Reassortment of La Crosse and Tahyna bunyaviruses in Aedes triseriatus mosquitoes. Virus Research. 20: 181-191. Conner, K., Sikora, E. J., Zhang, L. and Burmester, C. 2013. First report of soybean vein necrosis-associated virus affecting soybeans in Alabama. Plant Health Progress. 14. https://doi.org/10.1094/PHP-2013-0729-03-BR Escalante, C., Bollich, P. and Valverde, R. 2018. Soybean vein necrisos-associated virus naturally infecting yard-long bean (Vigna unguiculata ssp. Sesquipedalis) and soybean in Louisiana. Plant Disease. 102: 2047. Garnet, G. P. and Antia, R. 1994. Population biology of virus-host interactions. In: The evolutionary biology of viruses. Morse, S. (ed). Raven Press. New York, U.S. Han, J., Domier, L. L., Dorrance, A. E. and Qu, F. 2013. First report of Soybean vein necrosis- associated virus in Ohio soybean fields. Plant Disease. 97: 693. Hartman, G. L. 2015. Soybean diseases of importance in the United States. Pages 15-16 in: Compendium of soybean disease and pests. 5th edition. Hartman, G. L., Rupe, J. C., Sikora, E. J., Domier, L. L., Davis, J.A. and Steffey, K. L. ed. American Phytopathological Society, St. Paul, MN. Hsu, C. L., Hoepting C. A., Fuchs, M., Smith, E. A. and Nault, B. A. 2011. Sources of Iris yellow spot virus in New York. Plant Disease. 95: 735-743. Jacobs, J. L. and Chilvers, M.I. 2013. First report of Soybean vein necrosis virus on soybeans in Michigan. Plant Disease. 97: 1387. Keough, S. Han, J., Shuman, T., Wise, K. and Nachappa, P. 2016. Effects of soybean vein necrosis virus on life history and host preference of its vector, Neohydatothrips variabilis, and evaluation of vector status of Frankliniella tritici and Frankliniella fusca. Journal of Economic Entomology. 109: 1979-1987.

Kleczewski, N. 2016. Research updates on soybean vein necrosis virus. http://extension.udel.edu/fieldcropdisease/2016/01/20/research-updates-on-soybean-vein- necrosis-virus Loomans, A. J. M., Tolsma, J., Fransen, J. J. and van Lenteren, J. C. 2006. Releases of parasitoids (Ceranisus spp.) as biological control agents of western flower thrips (Frankliniella occidentalis) in experimental glasshouses. Bulletin of Insectology. 59: 85-97. Irizarry, M. D., Elmore, M. G., Batzer, J. C., Whitham, S. A. and Mueller, D. S. 2018. Alternative hosts of soybean vein necrosis virus and feeding preferences of its vector soybean thrips. Plant Health Progress. 19: 176-181. Margaria, P., Ciuffo, M., Rosa, C. and Turina, M. 2015. Evidence of a tomato spotted wilt virus resistance-breaking strain originated through natural reassortment between two evolutionary- distinct isolates. Virus Research. 196: 157-161. Ohnishi, J. Katsuzaki, H., Tsuda, S., Sakurai, T., Akutsu, K. and Murai, T. 2006. Frankliniella cephalica, a new vector of Tomato spotted wilt virus. Plant Disease. 90: 685. Palmieri, J. R. 1982. Be fair to parasites. Nature. 298: 220. Peterson, A. T., Carroll, D. S., Mills, J. N. and Johnson, K. M. 2004. Potential mammalian filovirus reservoirs. Emerging Infectious Disease. 10: 2073-2081. Premachandra, W. T., Borgemeister, C., Maiss, E., Knierim, D. and Poehling, H. M. 2005. Ceratothripoides claratris, a new vector of a Capsium chlorosis virus isolate infection tomato in Thailand. Phytopathology. 95: 659-663. Qiu, W. and Moyer, J. W. 1999. Tomato spotted wilt tospovirus adapts to the TSWV N gene- derived resistance by genome reassortment. Phytopathology. 89: 575-582. Riley, D. G., Joseph, S. V., Srinivasan, R., Stanley, D. 2011. Thrips vectors of tospoviruses. Journal of Integrated Pest Management. 1: 1-10. Smith, E. A., Ditommaso, A., Fuchs, M., Shelton, A. M. and Nault, B. A. 2011. Weed hosts for onion thrips (Thysanoptera: Thripidae) and their potential role in the epidemiology of Iris yellow spot virus in an onion ecosystem. Environmental Entomology. 40: 194-203. Smith, D. L., Fritz, C., Watson, Q., Willis, D. K., German, T. L., Phibbs, A., Mueller, D., Dittman, J. D., Saalau-Rojas, E. and Whitham, S. A. 2013. First report of Soybean vein necrosis disease caused by soybean vein necrosis-associated virus in Wisconsin and Iowa. Plant Disease. 97: 693. Tentchev, D., Verdin, E., Marchal, C., Jacquet, M., Aguilar, J. M. and Moury, B. 2011. Evolution and structure of Tomato spotted wilt virus populations: evidence of extensive reassortment and insight into emergence processes. General Virology. 92: 961-973. Tommasini, M. G. and Maini, S. 1995. Frankliniella occidentalis and other thrips harmful to vegetable and ornamental crops in Europe. Wageningen Agricultural University Papers. 95: 1- 42.

Tzanetakis, I. E., Wen, R.-H., Newman, M. 2009. Soybean vein necrosis virus: A new threat to soybean production in Southeastern United States? Phytopathology. 99: 131. Whitfield, A. E., Kumar, N. K. K., Rotenberg, D., Ullman, D. E., Wyman, E. A., Zietlow, C., Willis, D.K. and German, T.L. 2008. A soluble form of Tomato spotted wilt virus (TSWV) glycoprotein GN (GN-S) inhibits transmission of TSWV by Frankliniella occidentalis. Phytopathology. 98: 45-50. Yanase, T., Kato, T., Yamakawa, M., Takayoshi, K., Nakamura, K., Kokuba, T. and Tsuda, T. 2006. Genetic characterization of Batai virus indicates a genomic reassortment between orthobunyaviruses in nature. Archives of Virology. 151: 2252-2260. Yanase, T., Kato, T., Aizawa, M., Shuto, Y., Shirafuji, H., Yamakawa, M. and Tsuda, T. 2012. Genetic reassortment between Sathuperi and Shamonda viruses of the genus Orthobunyavirus in nature: implications for their genetic relationship to Schmallenberg virus. Archives of Virology. 157: 1611-1616. Yates, T. L., Mills, J. N., Parmenter, R. R., Ksiazek, T., Parmenter, C. A., Vande Castle, J. R. Calisher, C. H., Nichol, S. T., Abbott, K. D., Young, J. C., Morrison, M. L., Beaty, B. J., Dunnum, J. L., Baker, R. J., Salazar-Bravo, J. and Peters, C. J. 2002. The ecology and evolutionary history of an emergent disease: hantavirus pulmonary syndrome: evidence from two El Niño – driven precipitation, the initial catalyst of a trophic cascade that results in a delayed density – dependent rodent response, is sufficient to predict heightened risk for human contraction of hantavirus pulmonary syndrome. BioScience. 52: 989-998. Zhou, J., Kantartzi, S. K., Wen, R.-H., Newman, M., Hajimorad, M. R., Rupe, J. C., Tzanetakis, I. E. 2011. Molecular characterization of a new tospovirus infecting soybean. Virus Genes. 43: 289-295. Zhou, J. 2012. Master thesis. Characterization and epidemiology of Soybean vein necrosis associated virus. University of Arkansas, Fayetteville, U.S. Zhou, J. and Tzanetakis, I. E. 2013. Epidemiology of Soybean vein necrosis-associated virus. Phytopathology. 103: 966-971. Zhou, J., Aboughanem-Sabanadzovic, N., Sabanadzovic, S. and Tzanetakis, I. E. 2018. First report of soybean vein necrosis virus infecting kudzu (Pueraria montana) in the United States of America. Plant Disease. 102: 1674.

Chapter III

Effects of Mixed Infections of Soybean Vein Necrosis, Bean Pod Mottle and Soybean Mosaic Viruses in Soybean

Abstract Soybean vein necrosis virus (SVNV), the causal agent of the homonymous disease, has become the most prevalent virus infecting soybean in North America. Because of this fact there is a high likelihood of mixed infections between SVNV and other soybean viruses. Mixed virus infections alter disease symptoms, virus titer and epidemiology as well as virus localization in the plant. Soybean mosaic virus (SMV) and bean pod mottle virus (BPMV) are the most economically important viruses affecting soybean in the U.S. and are both widely distributed across the continent. In this study, we performed experiments to evaluate the interactions between SVNV, SMV and BPMV. We found that soybean, a local host for SVNV, becomes permissive in the presence of BPMV; whereas there were no interactions detected between

SVNV and SMV.

Introduction

Mixed virus infections is a common phenomenon in both natural and agricultural

ecosystems (Tollenaere et al., 2016). It has become an increasingly important research topic in

plant virology because of the effects in disease epidemiology and virus evolution. Furthermore,

understanding the molecular mechanisms and biological characteristics of mixed infections

provides insights into the pathogenesis of individual viruses, critical information for the

development of efficient and stable disease control strategies (Syller, 2012).

Mixed infections often generate distinct disease patterns and affect host vigor. For

example, umbraviruses cannot move between plants with an insect vector because they are

missing a coat protein. However, when co-infecting a plant with a compatible luteovirus, the

umbravirus genome is encapsidated in the luteovirus virion allowing for aphid transmission. On

the other hand, the luteovirus is able to break the phloem-barrier and move systemically (Syller,

2003; Ryabov et al., 2001). Other unpredictable pathological consequence of multiple virus infections include resistance breakdown. Tomato carrying the tomato spotted wilt virus Sw-5 resistance gene becomes susceptible when the virus is found in mixed infections with tomato chlorosis virus (García-Cano et al., 2006).

Soybean mosaic virus (SMV) and bean pod mottle virus (BPMV) are the most economically important viruses infecting soybean in the U.S. (Hartman, 2015). Yield losses could reach 35% for SMV (Stuckey et al. 1982) and 50% for BPMV (Hopkins and Mueller,

1984) whereas co-infections lead to severe stunting and diminished yield (Ross, 1969; Giesler et al., 2002; Hobbs et al., 2003). An additional effect is seed transmission; it occurs at low levels in single infections and increases in dual infections (Nam, et al. 2013). The prevalence of soybean vein necrosis virus (SVNV) in all major producing areas in North America (Zhou et al., 2011;

Ali and Abdalla, 2013; Conner et al., 2013; Han et al., 2013; Jacobs and Chilvers, 2013; Smith et

al., 2013; Kleczewski, 2016; Escalante et al., 2018) increases the likelihood of co-infections with

SMV and BPMV. Such interaction could add another layer of complexity to the diseases caused

by SMV and BPMV and alter the pathology and epidemiology of SVNV. The objective of this

study was to evaluate effects of mixed infections of SVNV, BPMV and SMV on soybean which

may change important biological and epidemiological characteristics of either virus.

Materials and Methods

Plant material and virus isolates

In preliminary studies, ‘Hutcheson’ exhibited typical disease symptoms upon sole

infections with SVNV, BPMV or SMV and therefore was used in this study. Seeds were

generously provided by soybean breeding program and Dr. John Rupe at the University of

Arkansas (Fayetteville, U.S.).

BPMV-infected soybean tissue was kindly provided by Dr. Sead Sabanadzovic at

Mississippi State University and SMV-G7 strain were provided by soybean breeding program at the University of Arkansas. Both viruses were maintained on ‘Hutcheson’ through continuous

mechanical back inoculations. Soybean leaves exhibiting typical vein necrosis symptoms were

collected from a field in Washington Country, Arkansas in September of 2016. The presence of

SVNV was confirmed using dot blot immunoassay and RT-PCR as previously described (Zhou and Tzanetakis, 2013; Zhou et al., 2018). Infected tissue was used as the source for SVNV and maintained on ‘Hutcheson’ using viruliferous Neohydatothrips variabilis. Briefly, adult thrips collected from the field were released to soybean seedlings grown under controlled conditions

(27°C, Light: Dark = 16h: 8h). Larvae (<48 h old) that hatched from the leaf epidermis were

pooled and fed on SVNV-infected tissue for 48 h to allow for virus acquisition. These larvae were released to 8-10 seedlings grown in a one-gallon pot in the growth chamber (27°C, Light:

Dark = 16h: 8h, 750 μmol·m-2·s-1). Typical virus symptoms became apparent at approximately

12-15 days post inoculation and the presence of SVNV was confirmed using dot blot immunoassay and RT-PCR as described (Zhou and Tzanetakis, 2013; Zhou et al., 2018).

Virus inoculation

Soybean plants at the unifoliate stage were mechanically inoculated with plant sap generated from virus-infected leaf tissue. Sap inoculum was prepared by grinding symptomatic tissue in fresh phosphate buffer (0.1 M, pH 7.2) containing 0.1% (vol/vol) 2-mercaptoethanol at

1:10 (wt/vol) ratio and inoculated using cotton balls onto seedlings kept in the dark overnight and dusted with carborundum (600 mesh). The mortar and pestle used for tissue grinding were precooled at -20°C for at least 1 h before use. For single virus inoculation, sap containing either

SVNV, BPMV or SMV was applied to each unifoliate leaves. For co-inoculation treatments, plant tissue infected with individual viruses was mixed 1:1 (wt/wt) before grinding. Plant sap containing SVNV+BPMV, SVNV+SMV or SVNV+BPMV+SMV was applied to each leaflet and the inoculated plants were maintained in growth chamber (20°C, Light: Dark = 16h: 8h) to allow for symptom development.

Nucleic acid extraction and reverse transcription

Total nucleic acids were extracted from trifoliate leaves of inoculated soybean plants as described previously (Zhou and Tzanetakis, 2013). To assess the quality of RNA extract, the universal NADH primers were used to amplify a 721-base region of NADH dehydrogenase ND-

2 subunit transcript (Tzanetakis et al., 2007; Thekke-Veetil and Tzanetakis, 2017). Reverse transcription (RT) was conducted in a 25 μl reaction consisting of 50 U of Maxima H Minus

reverse transcriptase, Maxima RT buffer, 6 U of RiboLock RNase inhibitor (all three are from

Thermo Fisher Scientific), 10 nmol virus-specific primer and 0.4 mM dNTPs, in addition of 10%

final volume of total DNA-free RNA. Virus specific primers used for BPMV and SMV were designed based on their coat protein genes (GenBank Accession Nos. GQ996947.1 and

U25673.1) (Table 1), respectively; whereas for SVNV, Sdet R was used as described (Zhou et al., 2011). Reactions were carried out in Thermo C-1000 (Thermo Fisher Scientific) with incubation at 50°C for 60 min followed by enzyme deactivation at 80°C for 5 min.

PCR

Synthesized cDNA (2.5 μl) was used as template in a 25 μl PCR reaction included 0.2 U

Taq polymerase (Genscript, NJ, U.S.), 10 × Taq buffer, virus detection primers and 0.4 mM dNTPs. Specific primers used for the amplification of individual viruses are listed in Table 1.

The PCR reaction for SVNV was initiated by denaturation at 94°C for 2min, followed by 40 cycles of denaturation at 94°C for 30 s, annealing at 55°C for 10 s, extension at 72°C for 30 s with a final extension at 72°C for 10 min. Similar conditions were adopted for SMV amplification, expect for a 58°C annealing temperature. For BPMV detection, cycles consisted of annealing at 60°C for 10s and extension for 15 s at 72°C. Amplification products were visualized in 1.5% agarose gel in 0.5 × TBE buffer stained with GelRed (Biotium, CA, U.S.), and the image was captured on GelDoc-It imaging system (UVP, CA, U.S.).

Dot blot immunoassay

Positive samples from RT-PCR were verified by dot blot immunoassay. Briefly, trifoliate leaflets were grinded in PBS buffer at 1:10 ratio (wt/vol) and centrifuged briefly to precipitate plant debris. The supernatant was dotted on GE Amersham Protran NC nitrocellulose membrane

(Thermo Fisher Scientific). The dotted membranes were dried and washed with PBS buffer

before transferred into blocking buffer (PBS, 0.1% Tween-20 and 5% non-fat milk) for 1 h

followed by washing with PBS-Tween (PBS buffer containing 0.1% Tween-20) and incubation

in antibody solution (Agdia Inc, IN, U.S.; 1:1000 dilution in PBS) for 1 h. The membrane was

then transferred to goat anti-rabbit alkaline phosphatase conjugate solution (Sigma, MO, U.S.;

1:5000 dilution in PBS containing 0.2% non-fat milk) and incubated for 1 h. After washing with

PBS-Tween, the membrane was transferred to substrate buffer (0.1 M Tris, pH 9.5, 0.1 M NaCl

and 5 mM MgCl2, containing 0.67% NBT and 0.33% BCIP). The reaction was stopped by

transferring the membrane to deionized water.

Results

‘Hutcheson’ is a restricted host for SVNV

The SVNV isolate used in this study was verified as restricted by ‘Hutcheson’ soybean as it does for other SVNV isolates (Zhou and Tzanetakis, 2013). Seedlings at the unifoliate stage

were mechanically inoculated with SVNV, infection and subsequent symptoms became evident

usually 7-10 days post inoculation as previously described (Zhou and Tzanetakis, 2013).

Inoculated leaflets exhibited chlorotic local lesions with the dimeter of 0.5-1.0 cm. Lesions on

the upper epidermis constituted of multiple smaller lesions of appropriately 0.1-0.5 mm which

initially appeared as red/purple without a halo (Fig 1). As disease progresses, the lesions

expanded as its central part became necrotic and usually bordered with purple halos which

gradually turned to yellow. Throughout the infection process, such symptoms were not observed

either on the non-lesion part of the same leaf blade or younger, non-inoculated leaves. The virus

was only detected from lesion areas but not in any other part of the inoculated plants using both

RT-PCR and dot blot assay, verifying previous results in ‘Hutcheson’ (Zhou and Tzanetakis,

2013).

‘Hutcheson’ is a permissive host for BPMV and SMV

Given that the isolates of these two viruses (BPMV and SMV) have never been used before on ‘Hutcheson’, we verified their ability to move systemically. When inoculated with

BPMV, soybean seedlings started showing symptoms including mild chlorotic mottling and mosaic on newly emerged leaves at appropriately 10-14 days post inoculation (Fig 2). The infection subsequently spread to other parts of the plant. Similarly, seedlings inoculated with

SMV did not display apparent virus-like symptoms on inoculated leaflets, however, mosaic pattern was evident on the first trifoliate and later stage leaves (Fig 3). Infected leaves had severe mosaic or blistered and distorted appearance as plant matured. BPMV and SMV were detected in both inoculated and non-inoculated leaves using RT-PCR and dot blot assay, verifying that the two viruses move systemically on ‘Hutcheson’.

SVNV moves systemically in ‘Hutcheson’ in the presence of BPMV but not SMV

BPMV- and SMV- systemic leaves were used in mechanical inoculations. For SVNV, growth chamber-grown plants infested with viruliferous thrips and exhibiting typical symptoms including vein clearing and chlorosis were tested for the virus. Those tested positive were collected and used as the inoculum source for the virus. In the course of the experiments, seedlings were mechanically inoculated with SVNV+BPMV, SVNV+SMV or

SVNV+BPMV+SMV. Tissues infected by a single virus were mixed at 1:1 (:1) ratio according to individual treatments.

Most of the double and triple inoculated plants did not show distinct symptoms on inoculated leaflets 14 day post inoculation, similar to single inoculation with BPMV or SMV.

Exceptions included a few SVNV-inoculated plants displaying purple local lesions. In

SVNV+BPMV and SVNV+BPMV+SMV, mild mottling or mosaic symptoms were observed on

younger, non-inoculated leaves 21 days post inoculation (Fig 4); whereas mosaic rather than

mottling was observed on SVNV+SMV. Symptom intensity did not increase as plants matured.

The trifoliate leaflets of all inoculated plants were collected at 21 days post inoculation and

tested by RT-PCR using the respective virus detection assays. The presence of SVNV was

detected in 8/80 and 11/84 plants of the SVNV+BPMV and SVNV+BPMV+SMV treatments,

respectively (Table 2). N no plant inoculated with SVNV+SMV was tested as positive for

SVNV. Amplification products (384 bp) corresponding to the SVNV NP gene were obtained

from direct sequencing and shared 96-99% identity with the reference isolate (GenBank Acc.

No. GU722319). Likewise, the infection of BPMV and SMV was also confirmed by obtaining

amplification products (276 bp for BPMV and 534 bp for SMV) sharing high identity (97-100%)

to their respective CP genes. Results obtained using RT-PCR were further confirmed by dot blot assays using antibodies generated against individual viruses. In the SVNV+BPMV+SMV group, individual plants were infected with either SVNV+BPMV, SMV+BPMV or all three viruses; none of the individuals was tested positive for SVNV+SMV simultaneously. This result along with the results of the dual infections showed that SVNV was always associated with BPMV but not SMV, suggesting that ‘Hutcheson’, a restricted host for SVNV, became permissive for infection by SVNV in the presence of BPMV.

Discussion

The impact of virus co-infection on plant disease dynamics has stimulated an increasing interest in further understanding its epidemiological and evolutionary outcomes. A growing number of reports on plants infected with multiple viruses suggest that the phenomenon is common in both wild plants and agricultural crops including soybean (Lamichhane and Venturi,

2015; Tollenaere et al., 2016). The widespread occurrence of SVNV in North America pointed to the potential of mixed infection between the virus and BPMV/SMV, the most economically important viruses for the U.S. soybean industry. Although SVNV was confined within local lesions in single infection, it was detected in systemic leaves in soybean plants co-infected with

BPMV.

This study was technically challenging. A reliable mechanical inoculation protocol for

SVNV was crucial before initiating studies that require uniform pathogen pressure. Previous studies revealed a low infection rate of the virus using mechanical inoculation on soybean (Zhou and Tzanetakis, 2013) and this occurred repeatedly during the initial stage of this study. On the other hand, inoculating SVNV using viruliferous thrips as the inoculum is not feasible. Virus transmission via thrips feeding is not uniform and screen cages did not adequately confine thrips to a single leaflet. Similar technical barriers were confronted while screening onion cultivars that are resistant to Iris yellow spot virus, another orthotopovirus (Bag et al., 2015). To circumvent this obstacle, efforts were made to reserve virion stability in the process of mechanical inoculation. Other than keeping mortars and pestles at -20°C for at least an hour before grinding, the number of plants on which each round of prepared inoculum was applied was limited to six.

Due to the pre-cooling step, the addition of inoculation buffer (routinely maintained at 4 °C and temporarily kept on ice before use) into mortars formed icy solution which may protect virus

particles from being damaged and further degraded by the heat generated during grinding. The inoculum was gently rubbed on unifoliate leaflets while the icy solution started melting whereas the completely melted inoculum was discarded.

Multi-virus infections often lead to loss of host tissue specificity for the participating viruses. Bean golden mosaic virus (BGMV), a phloem-limited begomovirus invaded non-phloem tissue in double infections with tobacco mosaic virus (TMV) (Carr and Kim, 1983). Likewise, another begomovirus - abutilon mosaic virus (AbMV) escaped the phloem when co-infected plants with cucumber mosaic virus (CMV) (Wege and Siegmund, 2007). The phenomenon also occurs in co-infections involving other viruses that are not phloem-limited. Zucchini yellow mosaic virus strain A facilitated the systemic movement of the M strain of cucumber mosaic virus, which is unable to move long-distance in the host (Cucurbita pepo; Choi et al., 2002). Bag et al. (2012) revealed that complementation between two orthotospoviruses, tomato spotted wilt virus (TSWV) and iris yellow spot virus (IYSV), allowed systemic movement of IYSV on

Datura (Datura stramonium), a non-permissive host for the virus.

As a relatively newly characterized orthotospovirus, SVNV infects soybean in a non- systemic manner. The virus has only been detected either within the local lesions generated through mechanical inoculation or the portion of leaflets where typical virus symptoms display due to viruliferous thrips feeding; no virus has been detected outside these areas (Zhou and

Tzanetakis, 2013). The fact that SVNV was detected in systemic leaves when co-inoculated with

BPMV is evidence that the latter facilitates SVNV to move out of the local lesions and infect leaves systemically. Such alternation in the movement of pattern of SVNV, however, was not observed on plants co-infected with SMV and SVNV.

Virus co-infections have been reported on soybean previously (Giesler et al., 2002;

Malapi-Nelson et al., 2009; Nam et al., 2013). The most prominent case is the BPMV - SMV coinfection where infected plants exhibited more severe disease symptoms compared to single infections (Giesler et al., 2002). The double infection causes enhancement of BPMV titer in infected plants and more importantly, drastic reductions in seed quality and yield (Calvert and

Ghabrial, 1983; Nam et al., 2013). Coinfection of SMV with alfalfa mosaic virus (AMV) resulted in reversed effects on the accumulation of the two viruses: the AMV titer spiked whereas SMV titer declined; and more severe disease symptoms were observed in comparison with single infections (Malapi-Nelson et al., 2009). All viruses involved in disease synergisms reported so far have genomes consist of positive sense, single-stranded RNA. Here, we present the first report of an interaction between positive sense- and ambisense- RNA viruses. SVNV is restricted to foliar tissues where the virus is introduced by thrips feeding. The fact that infection is always accompanied by localized symptoms indicates that soybean is probably not the preferred host of the virus from an evolutionary standpoint. In contrast, the mild disease symptoms observed on soybean plants co-infected with both viruses indicate that the presence of

BPMV may have improved the fitness of SVNV in soybean, making it a less virulent pathogen, which is crucial for establishing a long-term evolutionary relationship with a host (Childs and

Peters, 1993; Garnet and Antia, 1994; Yates et al., 2002).

The underlying reasons of the systemic movement of SVNV in the presence of BPMV remain to be understood. It is possible that the BPMV MP assists SVNV to move systemically.

Alternative mechanisms could include BPMV-coded protein(s) inhibiting the induction of host defense systems, eliminating the barriers for the long-distance movement of SVNV. Although the suppression of gene silencing activity has not been reported for BPMV, it has been shown

that the small coat protein (S-CP) of cowpea mosaic virus (CPMV), the type member of genus

Comovirus where BPMV belongs to, functions as a suppressor of posttranscriptional gene silencing (PTGS) (Liu et al., 2004).

Other than changing host tissue specificity and pathogen virulence, virus coinfection could potentially alter pathogen accumulation and vector-mediated transmission, the key components in disease epidemiology (Lamichhane and Venturi, 2015; Tollenaere et al., 2016).

The coinfection of southern rice black-streaked dwarf virus with rice ragged stunt virus caused not only the increased titers of both viruses compared with single infections but also increased virus acquisition efficiency of their corresponding vectors (Li et al., 2014). In contrast, the mixed infection of potato virus Y and potato leafroll virus, did not alter the titers of the viruses significantly; yet acquisition efficiency by green peach aphids (Myzus persicae Sulzer) significantly increased (Srinivasan and Alvarez, 2007). In terms of the interaction of SVNV and

BPMV, whether and how virus accumulation change and how that influences virus transmission needs to be determined experimentally. The effects of virus co-infections could be a species or even genotype-specific (González-Jara et al., 2004; Wintermantel et al., 2008; Tatineni et al.,

2010) and its consequential outcomes may be dependent on virus strains (Cassells and Herrich,

1977). Whether the systemic movement of SVNV observed on ‘Hutcheson’ in the presence of

BPMV is universal on other soybean genotypes as well as the effect of virus strain on the interaction need to be further studied. Last but not the least, timing of coinfection – whether the two viruses invade host plants simultaneously or sequentially is essential in estimating the impact of disease synergism (Goodman and Ross, 1974; Marchetto and Power, 2018). Due to the limited facilities and time frame of this project, only simultaneous infections of SVNV and

BPMV were performed. It is worth investigating virus coinfection under the conditions

mimicking the arrival timing of virus vectors in the field given that the two viruses are

transmitted by different vectors (thrips for SVNV and beetles for BPMV, respectively); which may not occur in the field at the same time.

Table 1. List of oligonucleotide primers used in RT-PCR.

Assay RT Primer Forward Primer Reverse Primer Size (bp)

SVNV GATTAAACAGAAAACTCCTTTG ACTTGTGCAAGCTTATGGT GAAATGATTCCAATCTGTTC 384

BPMV CTCCACCGAGAAGGTCAATAGA GGCTGATGGGTGTCCATATT CCACCGAGAAGGTCAATAGAAA 276 SMV TGTTTTGATTCACGTCCCTTG CCGCGTTTGCAGAAGATTAC AGCCTTCATCTGCGCTATT 534

Fig 1. SVNV local lesion on unifoliate leaflet generated by mechanical inoculation (left) with close-up (right).

Fig 2. Mottling symptoms of BPMV on newly emerged leaves.

Fig 3. Mosaic on systemic leaves caused by SMV.

Table 2. Results of mixed infection experiment. First column represents individual treatment; each treatment was performed in two individual experiments.

Experiments Inoculated Plants Positive Plants for SVNV Positive Plants for BPMV and /Experiment /Experiment SMV/Experiment

SVNV+BPMV 42/38 5/3 35/29 and 0/0 SVNV+SMV+BPMV 45/39 8/3 32/28 and 12/9 SVNV+SMV 41/37 0/0 0/0 and 17/19

Fig 4. Leaflets on soybean seedling co-infected with SVNV and BPMV (upper panel) and mock- inoculation group (lower panel). In each panel, picture on the left represents mechanically- inoculated unifoliate leaflets; the one on the right represents systemic leaves.

References

Ali, A. and Abdalla, O. A. 2013. First report of Soybean vein necrosis virus in soybean fields of Oklahoma. Plant Disease. 97: 1664. Bag, S., Mitter, N., Eid, S. and Pappu, H. R. 2012. Complementation between Two Tospoviruses Facilitates the Systemic Movement of a Plant Virus Silencing Suppressor in an Otherwise Restrictive Host. PLoS One. 7: e44803. Bag, S., Schwartz, H. F., Cramer, C. S., Havey, M. J. and Pappu, H. R. 2015. Iris yellow spot virus (Tospovirus: Bunyaviridae): from obscurity to research priority. Molecular Plant Patholgy. 16: 224-237. Calvert, L. A. and Ghabrial, S. A. 1983. Enhancement by soybean mosaic virus of bean pod mottle virus titer in doubly infected soybean. Phytopathology. 73: 993-997. Carr, R. J. and Kim, K. S. 1983. Evidence that bean golden mosaic virus invades non-phloem tissue in double infections with tobacco mosaic virus. Journal General Virology. 64: 2489-2492. Cassells, A. C. and Herrich, C. C. 1977. Cross protection between mild and severe strains of tobacco mosaic virus in doubly inoculated tomato plants. Virology. 78: 253-260. Childs, J. E. and Peters, C. J. 1993. Ecology and Epidemiology of arenaviruses. pp. 331-384. In: Salvato, M. S. (eds). The Arenaviridae. Plenum Press. New York, U.S.

Choi, S. K., Yoon, J. Y., Ryu, K. H., Choi, J. K., Palukaitis, P. and Park, W. M. 2002. Systemic movement of a movement-deficient strain of Cucumber mosaic virus in zucchini squash is facilitated by a cucurbit-infecting potyvirus. Journal of General Virology. 83: 3173-3178.

Conner, K., Sikora, E. J., Zhang, L. and Burmester, C. 2013. First report of soybean vein necrosis-associated virus affecting soybeans in Alabama. Plant Health Progress. 14. https://doi.org/10.1094/PHP-2013-0729-03-BR

de Oliveira, A. S., Bertran, A.G. M., Inoue-Nagata, A. K., Nagata, T., Kitajima, E. W., Oliveira Resende, R. 2011. An RNA-dependent RNA polymerase gene of a distinct Brazilian tospoviurs. Virus Genes. 43: 385-389.

de Oliveira, A. S., Melo, F. L., Inoue-Nagata, A. K., Nagata, T., Kitajima, E. W., Resende, R. O. 2012. Characterization of bean necrotic mosaic virus: a member of a novel evolutionary lineage within the Genus Tospovirus. PLoS One. 7: e38634.

Escalante, C., Bollich, P. and Valverde, R. 2018. Soybean vein necrisos-associated virus naturally infecting yard-long bean (Vigna unguiculata ssp. Sesquipedalis) and soybean in Louisiana. Plant Disease. 102: 2047.

García-Cano, E., Resende, R. O., Fernández-Muñoz, R., and Moriones, E. 2006. Synergistic interaction between Tomato chlorosis virus and Tomato spotted wilt virus results in breakdown of resistance in tomato. Phytopathology. 96: 1263-1269.

Garnet, G. P. and Antia, R. 1994. Population biology of virus-host interactions. In: The evolutionary biology of viruses. Morse, S. (ed). Raven Press. New York, U.S.

Giesler, L. J., Ghabrial, S., Hunt, T. E. and Hill, J. H. 2002. Bean pod mottle: A threat to U.S. soybean production. Plant Disease. 86: 1280–1289.

González-Jara, P., Tenllado, F., Martínez-García, B., Atencio, F. A., Barajas, D., Vargas, M., Díaz-Ruiz, J. and Díaz-Ruíz, J. R. 2004. Host-dependent differences during synergistic infection by Potyviruses with potato virus X. Molecular Plant Pathology. 5: 29-35.

Goodman, R. M. and Ross, A. F. 1974. Enhancement by potato virus Y of potato virus X synthesis in doubly infected tobacco depends on the timing of invasion by the viruses. Virology. 58: 263-271.

Han, J., Domier, L. L., Dorrance, A. E. and Qu, F. 2013. First report of Soybean vein necrosis- associated virus in Ohio soybean fields. Plant Disease. 97: 693.

Hartman, G. L. 2015. Soybean diseases of importance in the United States. Pages 15-16 in: Compendium of soybean disease and pests. 5th edition. Hartman, G. L., Rupe, J. C., Sikora, E. J., Domier, L. L., Davis, J.A. and Steffey, K. L. ed. American Phytopathological Society, St. Paul, MN.

Hobbs, H. A., Hartman, G. L., Wang, Y., Hill, C. B., Bernard, R. L., Pedersen, W.L. and Domier, L. L. 2003. Occurrence of seed coat mottling in soybean plants inoculated with Bean pod mottle virus and Soybean mosaic virus. Plant Disease. 87: 1333–1336.

Hopkins, J. D. and Mueller, A. J. 1984. Effect of Bean pod mottle virus on soybean yield. Journal of Economic Entomology. 77: 943–947.

Jacobs, J. L. and Chilvers, M.I. 2013. First report of Soybean vein necrosis virus on soybeans in Michigan. Plant Disease. 97: 1387.

Kleczewski, N. 2016. Research updates on soybean vein necrosis virus. http://extension.udel.edu/fieldcropdisease/2016/01/20/research-updates-on-soybean-vein- necrosis-virus

Lamichhane, J. R. and Venturi, V. 2015. Synergisms between microbial pathogens in plant disease complexes: a growing trend. Frontiers in Plant Science. 6: 385-396.

Leastro, M. O., Pallás, V., Resende, R. O., Sánchez-Navarro, J. A. 2017. The functional analysis of distinct tospovirus movement proteins (NSM) reveals different capabilities in tubile formation, cell-to-cell and systemic virus movement among the tospovirus species. Virus Research. 227: 57- 68.

Li, S., Wang, H. and Zhou, G. 2014. Synergism between southern Rice black-streaked dwarf virus and Rice ragged stunt virus enhances their insect vector acquisition. Phytopathology. 104: 794-799.

Liu, L., Grainger, J., Cañizares, M. C., Angell, S. M. and Lomonossoff, G. P. 2004. Cowpea mosaic virus RNA-1 acts as an amplicon whose effects can be counteracted by a RNA-2- encoded suppressor of silencing. Virology. 323: 37-48.

Malapi-Nelson, M., Wen, R.-H., Ownley, B. H., Hajimorad, M. R. 2009. Co-infection of Soybean mosaic virus and Alfalfa mosaic virus results in disease synergism and alteration in accumulation level of both viruses. Plant Dis. 93: 1259-1264.

Marchetto, K. M. and Power, A. G. 2018. Coinfection timeing drives host population dynamics through changes in virulence. The American Naturalist. 191: 173-183.

Nam, M., Bae, H., Hammond, J., Domier, L. L., Youn, Y., Lee, B. and Lim, H. 2013. Seed transmission rates of Bean pod mottle virus and Soybean mosaic virus in soybean may be affected by mixed infection or expression of the Kunitz Trypsin Inhibitor. Research in Plant Disease, 19, 114–117.

Ross, J. P. 1969. Effect of time and sequence of inoculations of soybeans with soybean mosaic and bean pod mottle viruses on yields and seed characters. Phytopathology. 59: 1404–1408.

Ryabov, E. V., Graser, G., Mayo, M. A., Barker, H. and Taliansky, M. 2001. Umbravirus gene expression helps Potato leafroll virus to invade mesophyll tissues and to be transmitted mechanically between plants. Virology. 286: 363-372.

Srinivasan, R. and Alvarez, J. M. 2007. Effect of mixed viral infection (Potato virus Y – Potato leafroll virus) on biology and preference of vectors Myzus persicae and Mascrosiphum euphorbiae (Hemiptera: Aphididae). Journal of Economic Entomology. 100: 646-655.

Stuckey, R. E., Ghabrial, S. A. and Reicosky, D. A. 1982. Increased incidence of Phomopsis sp. in seeds from soybeans infected with Bean pod mottle virus. Plant Disease. 66: 826–829.

Syller, J. 2003. Molecular and biological features of umbraviruses, the unusual plant viruses lacking genetic information for a capsid protein. Physiological and Molecular Plant Pathology. 63: 35-46.

Syller, J. 2012. Facilitative and antagonistic interactions between plant viruses in mixed infections. Molecular Plant Pathology. 13(2): 204-216.

Tatineni, S., Graybosch, R. A., Hein, G. L., Wequlo, S. N. and French, R. 2010. Wheat cultivar- specific disease synergism and alteration of virus accumulation during co-infection with Wheat streak mosaic virus and Triticum mosaic virus. Phytopathology. 100: 230-238.

Thekke-Veetil, T. and Tzanetakis, I. E. 2017. Development of reliable detection assays for blueberry mosaic- and blackberry vein banding-associated viruses based on their population structures. Journal of Virological Methods. 248: 191-194.

Tollenaere, C., Susi, H. and Laine, A-L. 2016. Evolutionary and epidemiological implications of multiple infection in plants. 2016. Trends in Plant Science. 21: 80-90.

Tzanetakis, I. E., Postman, J. D. and Martin, R. R. 2007. Identification, detection and transmission of a new vitivirus from menthe. Archive of Virology. 152: 2027-2033.

Wege, C and Siegmund, D. 2007. Synergism of a DNA and an RNA virus: enhanced tissue infiltration of the begomovirus Abutilon mosaic virus (AbMV) and mediated by Cucumber mosaic virus (CMV). Virology. 357: 10-28.

Yates, T. L., Mills, J. N., Parmenter, R. R., Ksiazek, T., Parmenter, C. A., Vande Castle, J. R. Calisher, C. H., Nichol, S. T., Abbott, K. D., Young, J. C., Morrison, M. L., Beaty, B. J., Dunnum, J. L., Baker, R. J., Salazar-Bravo, J. and Peters, C. J. 2002. The ecology and evolutionary history of an emergent disease: hantavirus pulmonary syndrome: evidence from two El Niño – driven precipitation, the initial catalyst of a trophic cascade that results in a delayed density – dependent rodent response, is sufficient to predict heightened risk for human contraction of hantavirus pulmonary syndrome. BioScience. 52: 989-998.

Zhou, J., Kantartzi, S. K., Wen, R.-H., Newman, M., Hajimorad, M. R., Rupe, J. C., Tzanetakis, I. E. 2011. Molecular characterization of a new tospovirus infecting soybean. Virus Genes. 43: 289-295.

Zhou, J. and Tzanetakis, I. E. 2013. Epidemiology of Soybean vein necrosis-associated virus. Phytopathology. 103: 966-971.

Chapter IV

Effects on the Transmission Efficiency of Soybean Vein Necrosis Virus Mediated by Site

Mutagenesis of the RGD29-31 and N229 Motifs

Abstract Orthotospoviruses are acquired by thrips during feeding on infected tissue. Virions travel

through the foregut and enter midgut epithelial cells through the interaction between the viral

glycoproteins and cellular receptors. The glycoproteins of many members in the genus

Orthotospovirus contain a RGD motif, characteristic of cell adhesion molecules and predicted to

serve as a receptor binding region. In addition, the N-linked glycosylation sites on the

glycoproteins may also play roles in virus cell entry. However, the function of those motifs in

orthotospovirus-thrips interactions have not been studied. The goal of this research is to tackle

the issue in the soybean vein necrosis virus (SVNV)/Neohydatothrips variabilis system. If the

RGD motif and the N- glycosylated residues are involved in virus-cell attachment, then feeding

thrips peptides containing those motifs could saturate cell receptors and block virion binding

sites. Fewer virus particles would be able to enter the midgut cells and virus transmission

efficiency would decrease. To test this hypothesis and further identify the key amino acid(s) that is (are) essential for virus transmission, we designed peptides containing target sequences and used alanine scanning to generate single, double and triple mutants. Results indicate that this strategy could significantly reduce the transmission efficiency of SVNV by Neohydatothrips variabilis.

Introduction

Thrips (order Thysanoptera) are insects feeding primarily on plants causing stippling, scarred and distorted appearances on the surface of leaves, flowers and fruit. They are the only known vectors of orthotospoviruses, viruses that cause significant economic losses on crops globally (Pappu et al., 2009; Oliver and Whitfield, 2016). There are more than 5500 thrips species described to date and 17, belonging to the genera Frankliniella, Thrips, Ceratothripoides,

Scirtothrips, Dictyothrips, Neohydatothrips and Taeniothrips have been confirmed as orthotospovirus vectors (Riley et al., 2011; Montero-Astúa, 2012; Zhou and Tzanetakis, 2013;

Xu et al., 2017). The interactions between tomato spotted wilt virus (TSWV) – the type member of the genus Orthotopovirus and its primary vector – Frankliniella occidentalis have been extensively studied and the results have shed light on many important attributes of thrips- orthotospovirus interactions (Rotenberg et al., 2015; Whitfield et al., 2015). Virus transmission occurs in a persistent propagative manner, transtadially but not transovarially. Although virions can be acquired by thrips throughout their lifetime, only the individuals that ingest viruses in the first and early second instars stages are capable of transmitting (Ullman et al., 2005; Whitfield et al., 2005).

Virions are acquired by thrips through piercing and sucking on epidermal and mesophyll cells of infected leaflets and subsequently travel from stylet through the foregut before reaching the midgut where virus replication occurs (Montero-Astúa, 2012; Badillo-Vargas, 2014). Several models have been proposed for virus movement in thrips; the commonly accepted one indicates that viruses enter midgut epithelium cells through receptor-mediated endocytosis which involves the interaction between viral glycoproteins and cellular receptors. Following entry, virus replication occurs in the cytoplasm. Virus progeny are released from infected cells through

shedding and move across midgut microvilli, basal surface of epithelial cells and muscle cells

surrounding basal membranes. Once virions leave the last membrane barrier of muscle cells, they

traverse basal membrane and microvilli of salivary gland, the critical organ in virus transmission,

and move with saliva into a canal leading to an efferent salivary and exit from the combined

salivary-food canal. New transmission event occurs when thrips eject viruses into another host

during feeding (Whitfield et al., 2005; Montero-Astúa, 2012; Badillo-Vargas, 2014).

Entry into host cells is the early step in virus infection process. Although the

internalization pathway of orthotospoviruses has not been fully characterized, by circumstantial

experimental evidence and analogy to other members in the order Bunyavirales, it is likely that

virions are engulfed into host cells through receptor-mediated endocytosis, a common entry

mechanism utilized by enveloped viruses (Whitfield et al., 2005). This process involves a series

of events; initiated by virion attachment to cellular receptors and culminate with the fusion

between viral and host membranes (Ullman et al., 2005; Marsh and Helenius, 2006). Studies

performed by Whitfield et al. (2004) noted that a synthesized, soluble form of TSWV

glycoprotein N (GN) binds to larval thrips guts and decreases virus acquisition, providing

evidence that GN protein is crucial in mediating the attachment of virion to receptors displayed on

the epithelial cells of the thrips midgut. GN is the N-terminal protein coded by the M RNA of

orthotospoviruses. GN along with the C-terminal glycoprotein (GC) play essential roles in virus

entry into host cells (Whitfield et al., 2005; Ullman et al., 2005). Sequence analysis of the

soybean vein necrosis virus (SVNV) GN protein revealed the presence of an RGD29-31 motif, characteristic of cell adhesion molecules. This motif binds specifically to intergrins, a large family of transmembrane proteins consist of non-covalently bound heterodimeric subunits through electrostatic interactions between the RGD residues and metal ions in integrins (Schwab

et al., 2013; Badillo-Vargas, 2014; Yu et al., 2014). In addition, the N-linked glycosylation sites on the GN protein (N25, N229 and N343 for SVNV) may also be involved in virus entry (Whitfield,

2004).

The GN protein is a good candidate for antiviral compounds given that its soluble form generated by Whitfield et al. (2004) binds thrips larval guts and inhibits TSWV transmission

(Whitfield et al., 2008). Other than synthetic proteins, utilizing peptides targeting key sequences or motifs of viral proteins is a promising strategy used to reduce virus transmission as demonstrated in both plant - and animal – virus interactions (Wild et al., 1992; Santos et al.,

2002; Firbas et al., 2006; Liu et al., 2010; Borrego et al., 2013; Muhamad et al., 2015; Yang et al., 2017).

The RGD motif is present in SVNV and several other orthotospoviruses belonging to the

American clade of the genus (Chen et al., 2013). Its function in virus attachment/entry to thrips midgut cells and the potential effect of this interaction on transmission of orthotospoviruses, however, is not well understood. We hypothesize that the RGD motif as well as the N-linked glycosylation site of the SVNV GN protein (N229 based on in-silico simulations) are critical in the early steps of virus infection process, and virus transmission efficiency will decrease when RGD- binding receptors are saturated with ligands prior to acquisition. To test this hypothesis, we developed peptides with single, double and triple amino acids mutations at the RGD29-31 site and

N229 using alanine scanning and evaluated their effects on transmission.

Materials and Methods

Neohydatothrips variabilis rearing

Adult thrips, Neohydatothrips variabilis, were collected from soybean fields and placed

on leaf dishes for 7-10 days to allow for oviposit. Petri dishes contained detached trifoliates from

soybean plants floated on water to serve as food source for thrips. Hatched larvae were

transferred to leaf dishes and reared to adults. They were then collected and released to soybean

seedlings grown in 20” by 10” planting trays. Trays were maintained in growth chambers under

controlled environment (27°C, Light: Dark = 16h: 8h) with seedlings renewed every two weeks.

Adults and larvae from the trays removed from the chambers were combined and placed onto the

new seedling trays as the source of thrips propagation; soybeans in these trays were tested using

ELISA and/or RT-PCR for SVNV as described previously (Zhou and Tzanetakis, 2013; Zhou et

al, 2018) to verify them as SVNV-free.

Peptides design and synthesis

In-silico analysis of the SVNV GN protein revealed the presence of signature motifs

present in orthotospovirus orthologs including a RGD29-31 domain, several putative N-linked

glycosylation sites (N25, 229, 343) and transmembrane domains (aa6–28, 317–339, 349–371) (Zhou et al.,

2011). The two putative glycosylated residues located within or in close proximity to

transmembrane domains (Asn25 and Asn343) were not included in downstream experiments as they are probably unavailable for binding based on in- silico analysis. Polypeptides were designed around the RGD29-31 domain and Asn229; and two sequences were selected for synthesis

based on the predicted solubility: NASIRGDHEVSQE25-37 and RLTGECNITKVSLTN215-229.

Peptides with single, double or triple mutations at RDG29-31 and N229 are described in Table 1.

All peptides were synthesized at GenScript (NJ, U.S.; >95% purity).

Peptide delivery

Peptides RGD, AGD, RGA, AGA, RAA, AAA and N229 were dissolved in H2O

(Molecular Biology Grade, Quality Biological, MD, U.S.) to a 10 mM stock solution; peptides

RAD and AAD were dissolved in water containing 28-30% NH4OH as they were insoluble in

H2O; A229 did not dissolve in either water or water containing up to 30% NH4OH (as suggested

by the manufacturer).

Before used in downstream experiments, the peptide stock solution was diluted in feeding

buffer (8% sucrose, 20% blue food dye). One hundred and sixty (160) µl of 10 nM peptide(s)

were pipetted inside the lid of a 1.5 µl Eppendorf tube with the lid being sealed with a piece of

stretched parafilm (2 cm × 2 cm before stretch). First instar larvae, hatched within 24 h of the

initiation of the feeding experiments were collected from SVNV-free leaf dishes and individually transferred to the bottom of the tube using size 0 camel painting brush. The behavior of each larvae was observed through a dissecting microscope for at least 10 seconds and only individuals with good vitality, exhibited as active movement along the vial wall were kept within the tube before closing the lid. These tubes were then placed in a black-colored microtube rack (VWR,

PA, U.S.) under a light source in order to attract larvae to move upward the vial and feed.

Peptide solution confined within the lid of tubes is acquired by larvae through feeding on the stretched parafilm. To analyze the potential effect of different types of peptides on transmission efficiency, treatments were set up as listed in Table 2.

A cohort of 1st instar larvae (5-6) were fed on peptide solution as such in each tube for 5 h

and then transferred onto SVNV-infected soybean leaf tissue for virus acquisition. Soybean

leaflets showing typical SVNV symptoms (vein-clearing and chlorosis) and tested positive for

SVNV by RT-PCR were used as virus source. Briefly, after peptide acquisition, larvae were

individually transferred onto SVNV-infected tissue floating on water in a petri dish. After a 16 h

- acquisition access period (AAP), larvae were transferred to a SVNV-free leaf dish. The petiole

of the leaflet was dipped into 1 ml water to keep the leaf alive. After larvae were transferred onto

the leaf, the petri dish was sealed with parafilm to prevent escape. Larvae were reared on such

leaf dish until the 2nd larvae stage. Control groups were a) larvae fed on feeding buffer and b)

larvae fed on feeding buffer before exposed to the virus. Additionally, the accessibility of the

virus in infected tissue used in each experiment was tested by feeding larvae solely on that piece

of tissue without getting access to either the peptide solution or feeding buffer.

Peptide effects on transmission

Individual 2nd instar larvae were transferred from the leaf dish to a leaf blade of a soybean

seedling at unifoliate stage growing in a 3 oz pot. To prevent larvae from escaping, the pot was

covered with a cage made from a 9 oz clear plastic cup with its bottom removed and replaced by

a piece of thrips-proof mesh (8 cm × 8 cm) glued from outside to allow ventilation. The caged

seedlings were grown in the growth chamber at 27 °C with a 16 h photoperiod. Larvae were

allowed to feed on plants, develop to adults and propagate within cages; 20 d post transfer, cages

were removed and plants were screened for SVNV infection using dot-blot immunoassay and

RT-PCR as previously described (Zhou and Tzanetakis, 2013; Zhou et al, 2018). The data was analyzed using one-way analysis of variance (ANOVA) on percentage of virus infection and percentage of thrips feeding using JMP Pro 13 (SAS Institute Inc., Cary, NC).

Peptide sequencing

To further confirm sequences of peptides RGD and AAA, those peptides were analyzed

using MALDI-TOF-MS and LC-ESI-MS/MS at the Mass Spectrometry Facility Center of

University of Arkansas (Fayetteville, U.S.) as previously described (Caprioli et al., 1997; Dams,

et al., 2003).

Results

Peptide delivery

A homogeneous blue color observed through the abdomen of the larvae after feeding with peptide solution was used as an indicator of successful peptide delivery (Fig 1) and only blue larvae were harvested and used for downstream experiments.

Transmission

Four feeding groups (peptide/SVNV, buffer/SVNV, buffer, SVNV) including nine treatments (Table 2) were analyzed for thrips ability to transmit the virus. Dot blot results showed a perfect correlation between typical SVNV symptoms (vein-clearing and chlorosis) and local lesions on leaf surface (Zhou and Tzanetakis, 2018). For plants that only exhibit local lesions, RT-PCR was used to detect the presence of SVNV given that the amount of symptomatic leaf tissue may have not been enough for dot blot detection (Zhou and Tzanetakis,

2013). Stippling, scars and distorted appearances on plant surface verify thrips feeding. Plants lacking such signs were excluded from the analysis. Each treatment was repeated in three experiments with multiple thrips tested. The ratio of number of plants fed by thrips to total number of plants transferred with thrips and the ratio of number of plants infected with SVNV to number of plants fed by thrips were calculated for each replicate, respectively, together with their percentages (Table 2). In addition to serve as the technical control of the experiment, SVNV treatment (where thrips were exposed to the virus) also measured the transmission efficiency of single thrips.

Statistical analysis

SVNV infection (Fig 2) and thrips feeding rates (Fig 3) were analyzed using One-way

ANOVA. Analysis of infection rate reveals a significant difference between treatments (P <

0.0001). The post hoc test using Dunnett’s method suggests that treatments fed with wild-type peptides (RGD + N229), single-mutation peptides combination (AGD + RAD + RGA) and triple-

mutation peptide (AAA) have significantly lower infection rates (P < 0.05) when compared with

the control group (thrips were fed with buffer prior to the exposure to SVNV). Treatments fed

with double-mutation peptides combination (AAD + AGA + RAA), RGD29-31 or N229 did not

have pronounced difference (P > 0.05; Table 4). In addition, the infection rate of control group

does not differ significantly from the group fed on SVNV tissue. On the other hand, ANOVA did

not reveal a significant difference on feeding rate across treatments (P = 0.512; Table 5) and

there is weak correlation between feeding and infection rates (R = 0.1463).

Peptide sequencing

Analysis of peptides RDG and AAA using MALDI-TOF-MS revealed single peaks at

1441.9 KD and 1326.6 KD, respectively. These values correspond to their individual molecular

mass provided by GenScript (NJ, U. S.) indicating their intact masses. Moreover, their sequences

were further analyzed by LC-ESI-MS/MS; results showed expected sequences as designed (Fig

4).

Discussion

The usage of synthetic peptides as therapeutic agents has been explored in treating animal

and human viral diseases (Houimel and Dellagi, 2009; Arosio et al., 2012; Gosselet et al., 2013;

Hipolito et al., 2014; Muhamad et al., 2015; Zhang et al, 2015) whereas their application in

controlling plant-infecting viruses is still in its infancy. Peptides that confer disease resistance are

primarily interfering with viral proteins either associated with replication such as the

nucleoprotein of orthotospoviruses and the replicase of geminivirus (Rudolph et al., 2003;

Lopez-Ochoa et al., 2006) or virion assembly as seen with luteoviruses (Liu et al., 2010). The

potential of utilizing peptides to block viral proteins mediating attachment and entry is yet to be

studied.

It has been shown that GN protein of TSWV is the candidate target of antiviral

compound (Whitfield et al., 2004; Whitfield et al., 2008) and the cellular adhesion hallmark –

RGD motif is predicted as critical in virion attachment. The role of this motif in orthotospovirus

infection process and whether it could serve as a potential target of antiviral compound is

unclear. The RGD motif has been identified in GN proteins of several members in the genus of

Orthotospovirus including SVNV, a distinct species transmitted by an uncommon vector N. variabilis. This special orthotospovirus - thrips interaction may reflect unique transmission properties that are critical for the co-evolution of orthotopoviruses with their vectors. This study designed polypeptides containing sequences of interest including RGD29-31 sequence, Asn229 and

the mutated forms of these sequences. By comparing transmission efficiency of N. variabilis fed

with either single peptide or peptide combinations prior to virus acquisition, we found intriguing

results which suggest peptides derived from RGD motif could be useful to decrease transmission

rate of SVNV.

Those experiments also pointed out to the high SVNV transmission efficiency with single

thrips (above 36%; Table 2). Because of that and in order to control virus dispersal, it is

imperative to develop strategies that could block virion entry into thrips midgut epithelium cells.

Transmission rates did not differ between thrips fed with buffer prior to SVNV acquisition and

those fed only with SVNV (P = 1.00) indicating that the feeding buffer does not interfere with

virus transmission. The addition of blue food dye in the buffer not only attracts larvae to feed on

solutions containing polypeptides but also enables us to visualize the acquisition status of the

mixture. Survival rates of larvae did not differ from groups fed with and without buffer

suggesting the nontoxicity of the buffer. In the presence of RGD and Asn229 peptides,

transmission was greatly reduced (P < 0.0001). When thrips were fed with either RGD (P = 0.72)

or Asn229 - (P = 0.18) peptides separately, there was no significant reduction in transmission

compared to the controls. These results indicate that both the RGD and Asn229 motifs are needed

for successful cell entry and therefore transmission. To test this hypothesis, a peptide was synthesized where Asn229 was substituted by Ala. The insolubility of this peptide, however,

prevented further evaluation of the role of Asn229 in virus transmission.

Evaluating the transmission rate of larvae fed with a combination of peptides consisting

of different mutated RGD sequences is the preliminary step to identify peptide(s) with key amino

acid(s) in virus transmission. Some peptides lost their binding activity to receptors due to the key

amino acid(s) for such activity was substituted with Ala, whereas some still retained this function

because Ala replaced non-essential amino acid(s). Peptides combination consist of single-amino

acid mutations including AGD, RAD and RGA exhibit inhibitory effect on virus transmission

when compared with control group (Table 3). However, such inhibitory effects are less potent

than the group fed with peptides containing RGD29-31 and Asn229 (Table 3). A possible

explanation is that single-amino acid mutants bind with cellular receptors but with either weaker

strength or coverage, which lead to partial blockage of the receptors. Such suppression on virus

transmission was further diminished by replacing two amino acids in RGD with Ala residues: the

transmission rate of double-mutants group containing sequences of AAD, AGA and RAA is not

significantly lower than control group (Table 3). These results follow the line of logic that RGD

motif is essential in virion-cell attachment and the disruption of this region impairs host’s ability

to transmit viruses.

The contribution of individual amino acids consisting RGD motif in virus transmissibility

seems to vary among different pathosystems, and the amino acid residue utilized to generate

mutants may affect the results of mutagenesis studies. In adenovirus type 2 where RGD340-342 of

the penton base protein, substitution of Arg340->Ser, Gly341->Val and Asp342 ->Glu abolished

virus activity respectively suggesting each amino acid is crucial for the infection process (Bai, et

al., 1993). On the other hand, Wei et al. (2014) found Arg and Gly but not the Asp were essential

for fusion activity for Human metapneunovirus. In the case of foot-and-mouth disease virus

(FMDV), its virulence is abolished by replacing Asp143->Ala of the RGD141-143 motif but not by

mutants targeting the other two residues (Gutiérrez-Rivas et al., 2008).

Two peptide treatments produced intriguing results and for this reason the molecular

masses and amino acid sequences of RGD and AAA were confirmed by MALDI-TOF-MS and

LC-MS/MS to exclude the possibility of contamination (Fig 4 and 5). The transmission rate of

RGD is not significantly lower than the controls and unlike RGD/Asn229 (Table 3), indicating

that probably both RGD29-31/Asn229 motifs are needed for efficient virus entry. The transmission

rate of AAA is comparable to that of RGD/Asn229 (Table 3). It is possible that AAA blocks the

cellular receptor, leading to transmission suppression. By reducing the load of virus particles on

epithelium cells, the transmission efficiency diminishes. This hypothesis is supported by the fact

that the myristoylated alanine-rich C-kinase substrate (MARCKS) protein inhibits cellular

adhesion to extracellular matrix proteins including fibronectin (Spizz and Blackshear, 2001), a

phenomenon that could also occur in the presence the RGD motif, which is a known fibronectin

competitive inhibitor (Pierschbacher and Ruoslahti, 1984). Alanine-rich protein/peptides could inhibit cellular binding not only to endogenous integrin-binding proteins that contain the RGD motif but also RGD-containing viral proteins, such as the SVNV GN protein, leading to the reduction of virus transmission. Moreover, given the fact that MARCKS inhibition of cell adhesion is independent of direct integrin receptor modulation (Spizz and Blackshear, 2001), alanine-rich proteins/peptides may be promising in impeding virus entry regardless of the type of host cellular receptors.

The observed results could also be due to in vivo factors that influence the binding of the

RGD motif. The microenvironment of virion-cell attachment consists of several components including 1) host cells with integrin incorporated on their surface, 2) viruses which recognize cells through the integrin-binding domains of the virus glycoproteins and 3) native integrin- binding proteins which do not exist in in vitro experiments and may have not been identified in in vivo studies. The endogenous integrin-binding proteins known as fibronectin, fibrinogen and vitronectin and other glycoproteins are present in blood and other body fluids of vertebrates at high concentrations (Bellis, 2011; Schwab et al., 2013). Despite of the lack of information on their presence in thrips, as the essential players of the evolutionary conserved cell-adhesion systems, these integrin receptors have also been identified in different species of invertebrates including arthropods (Akiyama and Johnson, 1983; Pradel et al., 2004; Hanington and Zhang,

2010). The binding signal triggered by these proteins to integrins on host cell surface is more robust than exogenous factors (Woods et al., 1986; Aota et al., 1991). The addition of AAA- containing peptide may have stimulated the secretion of endogenous integrin-binding proteins through interacting with proteins involved in related molecular pathways, similar to peptides interfering with protein-protein interaction in the ethylene signaling pathway (Bisson et al.,

2016), causing saturation of integrin-binding sites and consequently blocking the attachment of

virions to host cells.

On the other hand, the results obtained from RGD and AAA may indicate that the

interaction between SVNV and N. variabilis could involve non-RDG components. This

hypothesis is supported by the fact that among all characterized orthotospoviruses, only members

belonging to American clade contain a RGD motifs in their respective GN proteins (Chen et al.,

2013). Molecular signals mediating cell-virion attachment in Euroasian clade have not been

identified to date. It is very likely these viruses utilize RGD-independent mechanisms such as lectin-like domain identified in TSWV GN protein to get access to host cells (Whitfield, 2004).

The interplay between RGD-containing molecules and its cellular receptors is one of the most extensively studied protein-protein interactions given the ubiquitous presence of RGD motif in various extracellular matrix proteins (Mecham, 2011; Zapp et al., 2018). The potential of using RGD-containing peptides to block virus attachment has been investigated for animal-

and human-infecting viruses including adenoviruses, foot-and-mouth disease virus (FMDV) and

human immunodeficiency virus -1 (HIV-1) and in-vitro studies have shown a reduction of virus intake in the presence of synthetic peptides (Bai et al., 1993; Santos, et al, 2002; Borrego et al.,

2013). However, the function of RGD motif in the transmission of plant-infecting viruses has only been inferred by analogy to related human or animal viruses; direct experimental evidence is not available.

Despite the complicated in vivo studies, we showed the possibility of using RGD- containing peptides to decrease virus transmission. Whether the reduction in transmission efficiency is due to peptide-blockage of cellular receptor needs to be elucidated. Although the amino acid composition of RGD is crucial in integrin-mediated cell attachment, it is also

essential that viral glycoprotein fold correctly so the motif is exposed at the surface of the

protein, making it accessible to its receptors. On the other hand, the stereochemistry of RGD

peptide also influences the interaction between integrin receptors and the conserved motif.

Amino acid substitution in the flanking sequences could alter binding specificity and strength of

the conserved motif to its receptors (Pierschbacher and Ruoslahti, 1987; Plow et al., 1987; Liu et

al., 1994; Haubner et al., 1996). Likewise, three-dimensional peptide conformations have similar

effects; for example the linear RGD sequences such as GRGDSP and RGDSPASSKP bind

preferentially to α5β1, whereas their cyclic counterparts GPenGRGDSPCA and cyclo (RGDf

(NMe)V) are selective for αvβ3 (Hersel et al., 2003). Additionally, it is also important RGD

sequence is presented in a context that can be recognized by integrin and compatible with its

binding (Bellis, 2011).

The transmission efficiency of SVNV by individual N. variabilis is over 36%,

underlining the importance of disrupting virus-thrips interaction for effective virus control. The analysis of transmission efficiency mediated by synthetic polypeptides revealed that peptides derived from RGD29-31 and N229 motifs located at viral glycoprotein had the potential of blocking

virion attachment to cellular receptors and consequently decreasing virus transmission

efficiency. Future studies should focus on in vitro binding of selected peptides to thrips midgut

epithelium cells. There needs to be a clear understanding of how these motifs function in virion-

cell attachment evidence from both in vivo and in vitro assays.

Table 1. Name and sequence of synthesized peptides. Mutated amino acids are highlighted in bold.

Peptide Name Mutation Status Amino Acid Sequence

RGD Wild type NASIRGDHEVSQE N229 Wild type RLTGECNITKVSLTN AGD Single mutation NASIAGDHEVSQE RAD Single mutation NASIRADHEVSQE RGA Single mutation NASIRGAHEVSQE A229 Single mutation RLTGECAITKVSLTA AAD Double mutation NASIAADHEVSQE AGA Double mutation NASIAGAHEVSQE RAA Double mutation NASIRAAHEVSQE AAA Triple mutation NASIAAAHEVSQE

Fig 1. First instar larvae that has successfully acquired (first row) or failed to acquire peptide solution (second row).

Table 2. Transmission inhibition assays. Feeding: number of plants with feeding scars/plants exposed to thrips; Infection: number of SVNV–infected plants/plants with feeding scars. The numbers in parenthesis indicate the percentile of feeding and infection for each treatment.

Experiment 1 Experiment 2 Experiment 3 Total Treatment Feeding Infection Feeding Infection Feeding Infection Feeding Infection RGD + N229/SVNV 60/88 10/60 69/76 7/69 53/92 5/53 182/256 22/182 (68.2%) (16.7%) (90.8%) (10.1%) (57.6%) (9.4%) (71.1%) (12.1%) AGD + RAD + RGA/SVNV 53/80 13/53 42/60 9/42 89/101 22/89 184/241 44/184 (66.3%) (24.5%) (70.0%) (21.4%) (88.1%) (24.7%) (76.3%) (23.9%) AAD + AGA + RAA/SVNV 73/100 23/73 71/80 25/71 76/80 21/76 220/260 69/220 (73%) (31.5%) (88.8%) (35.2%) (95%) (27.6%) (84.6%) (31.4%) AAA/SVNV 26/100 5/26 65/120 6/65 58/60 7/58 149/280 18/149 (26%) (19.2%) (54.2%) (9.2%) (96.7%) (12.1%) (53.2%) (12.1%) RGD/SVNV 17/45 4/17 23/30 7/23 93/160 40/93 133/235 51/133 (37.8%) (23.5%) (76.7%) (30.4%) (58.1%) (43.0%) (56.6%) (38.3%) 87

N229/SVNV 37/44 7/37 20/23 7/20 116/154 35/116 173/231 49/173 (84.1%) (18.9%) (87.0%) (35%) (75.3%) (30.2%) (74.9%) (28.3%) Buffer/SVNV 38/56 14/38 63/85 23/63 95/105 38/95 196/246 75/196 (67.9%) (36.8%) (74.1%) (36.5%) (90.5%) (40.0%) (79.6%) (38.2%) SVNV 35/46 13/35 43/55 18/43 74/133 24/74 152/234 55/152 (76.1%) (37.1%) (78.2%) (41.9%) (55.6%) (32.4%) (65.0%) (36.2%) Buffer 42/63 0/42 68/93 0/68 54/70 0/54 164/226 0/164 (66.7%) (0%) (73.1%) (0%) (77.1%) (0%) (72.6%) (0%)

Fig 2. Infection rate of different treatments.

Table 3. Comparison of infection rates (mean ± standard error) among treatments.

Treatment P-value Infection rate (mean ± standard deviation)

RGD + N229/SVNV < 0.0001 12.07 ± 4.03 AGD + RAD + RGA/SVNV 0.02 23.53 ± 1.85 AAD + AGA + RAA/SVNV 0.58 31.43 ± 3.80 AAA/SVNV 0.0002 13.50 ± 5.14 RGD/SVNV 0.72 32.30 ± 9.89

N229/SVNV 0.18 28.03 ± 8.27 Buffer/SVNV (Control) 1.00 37.77 ± 1.94 SVNV 1.00 37.13 ± 4.75 Buffer < 0.0001 0

Fig 3. Feeding rate of different treatments.

Table 4. Comparison of feeding rates (mean ± standard error) among treatments.

Treatment P-value Feeding rate (mean ± standard deviation)

RGD + N229/SVNV 1.00 72.2 ± 16.96 AGD + RAD + RGA/SVNV 1.00 74.8 ± 11.67 AAD + AGA + RAA/SVNV 0.99 85.6 ± 11.34 AAA/SVNV 0.66 59.0 ± 35.59 RGD/SVNV 0.59 57.5 ± 19.46

N229/SVNV 1.00 82.1 ± 6.09 Buffer/SVNV 1.00 77.5 ± 11.68 SVNV 0.99 70.0 ± 7.21 Buffer 1.00 72.3 ± 5.25

Fig 4. Analysis of molecular masses and amino acid sequences of peptides RGD and AAA using MALDI-TOF-MS.

Fig 5. Analysis of amino acid sequences of peptides RGD (upper) and AAA (lower) using LC- ESI-MS/MS.

References

Akiyama, S. K. and Johnson, M. D. 1983. Fibronectin in evolution: presence in invertebrates and isolation from Microciona prolifera. Comparative Biochemistry and Physiology. B. 76: 687-694.

Aota, S., Nagai, T. and Yamada, K. M. 1991. Characterization of regions of fibronectin besides the arginine-glycine-aspartic acid sequence required for adhesive function of the cell-binding domain using site-directed mutagenesis. The Journal Biological Chemistry. 266: 15938e43.

Arosio, D., Casagrande, C. and Manzoni, L. 2012. Integrin-mediated drug delivery in cancer and cardiovascular diseases with peptide-functionalized nanoparticles. Current Medicinal Chemistry. 19: 3128-3151.

Bai, M., Harfe, B. and Freimuth, P. 1993. Mutations the alter the Arg-Gly-Asp (RGD) sequence in the adenovirus type 2 penton base protein abolish its cell-rounding activity and delay virus reproduction in flat cells. Journal of Virology. 67: 5198-5205.

Badillo-Vargas, I. E. 2014. Dissecting the molecular interplay between Tomato spotted wilt virus and the insect vector, Frankliniella occidentalis. Kansas State University, Manhattan, U. S.

Bellis, S. L. 2011. Adventages of RGD peptides for direction cell association with biomaterials. Biomaterials. 32: 4205-4210.

Bisson, M. M., Kessenbrock, M., Müller, L., Hofmann, A., Schmitz, F., Cristescu, S. M. and Groth, G. Peptides interfering with protein-protein interactions in the ethylene signaling pathway delay tomato fruit ripening. Scientific Reports. 6: 30634.

Borrego, P., Calado, R., Marcelino, J. M., Pereira, P., Quintas, A., Barroso, H. and Taveira, N. 2013. An ancestral HIV-2/simian immunodeficiency virus peptide with potent HIV-1 and HIV-2 fusion inhibitor activity. AIDS. 27: 1081-1090.

Caprioli, R. M., Farmer, T. B. and Gile, J. 1997. Molecular imaging of biological samples: localization of peptides and proteins using MALDI-TOF MS. Analytical Chemistry. 69: 4751- 4760.

Chen, T. C., Li, J.T., Fan, Y. S., Yeh Y. C., Yeh, S. D. and Kormelink, R. 2013 Molecular characterization of the full-length L and M RNAs of Tomato yellow ring virus, a member of the genus Tospovirus. Virus Genes. 46: 487-495.

Dams, R., Huestis, M. A., Lambert, W. E. and Murphy, C. M. 2003. Matrix effect in bio-analysis of illicit drugs with LC-MS/MS: influence of ionization type, sample preparation, and biofluid. Journal of the American Society for Mass Spectrometry. 14: 1290-1294.

Firbas, C., Jilma, B., Tauber, E., Buerger, V., Jelovcan, S., Lingnau, K., Buschle, M., Frisch, J. and Klade, C. S. 2006. Immunogenicity and safety of a novel therapeutic hepatitis C virus

(HCV) peptide vaccine: a randomized, placebo controlled trial for dose optimization in 128 healthy subjects. Vaccine. 24: 4343-4353.

Gosselet, F., Saint-Pol, J., Candela, P and Fenart, L. 2013. Amyloid-beta peptides, Alzheimer’s disease and the bold-brain barrier. Current Alzheimer Research. 10: 1015-1033.

Gutiérrez-Rivas, M., Pulido, M. R., Baranowski, E., Sobrino, F. and Sáiz, M. Tolerance to mutations in the foot-and mouth disease virus integrin-binding RGD region is different in cultured cells and in vivo and depends on the capsid sequence context. Journal of General Virology. 89: 2531-2539.

Hanington, P. C. and Zhang, S-M. 2010. The primary role of fibrinogen-related proteins in invertebrates is defense, not coagulation. Journal of Innate Immunity. 3: 17-27.

Haubner, R., Gratias, R., Diefenbach, B., Goodman, S. L., Jonczyk, A. and Kessler, H. 1996. Structural and functional aspects of RGD-containing cyclic pentapeptides as highly potent and selective integrin alphavbeta3 antagonists. Journal of the American Chemical Society. 118: 7461 -7472.

Hersel, U., Dahmen, C. and Kessler, H. 2003. RGD modified polymers: biomaterials for stimulated cell adhesion and beyond. Biomaterials. 24: 4385-4415.

Hipolito, S. G., Shitara, A., Kondo, H. and Hirono, I. 2014. Role of Marsupenaeus japonicas crustin-like peptide against Vibrio penaeicida and white spot syndrome virus infection. Developmental and Comparative Immunology. 46: 461-469.

Houimel, M. and Dellagi, K. 2009. Peptide mimotopes of rabies virus glycoprotein with immunogenic activity. Vaccine. 27: 4648-4655.

Lin, H. B., Sun, W., Mosher, D. F., Garcia-Echeverria, C., Schaufelberger, K., Lelkes, P. I. and Cooper, S. L. 1994. Synthesis, surface, and cell-adhesion properties of polyurethanes containing covalently grafted RGD-peptides. Journal of Biomedical Materials Research. 28: 329 - 342.

Liu, S., Sivakumar, S., Sparks, W. O., Miller, W. A. and Bonning, B. C. 2010. A peptide that binds the pea aphid gut impedes entry of Pea enation mosaic virus into the aphid hemocoel. Virology. 401: 107-116.

Lopez-Ochoa, L., Ramirez-Prado, J. and Hanley-Bowdoin, L. Peptide aptamers that bind to a germinivirus replication protein interfere with viral replication in plant cells. Journal of Virology. 80: 5841-5853.

March, M. and Helenius, A. 2006. Virus entry: open sesame. 2006. Cell. 124: 729-740.

Mecham, R. 2011. The Extracellular Matrix: an Overview. Berlin: Springer. doi: 10.1007/978-3- 642-16555-9.

Montero-Astúa, M. 2012. Ph.D dissertation. Unveiling and blocking the interaction between Tomato spotted wilt virus and its insect vector, Frankliniella occidentalis. Kansas State University, Manhattan, U. S.

Muhamad, A., Ho, K. L., Basyaruddin Abdul Rahman, A., Tejo, B. A., Uhrin, D. and Tan, W. S. 2015. Hepatitis B virus peptide inhibitors: solution structures and interactions with the viral capsid. Organic and Biomolecular Chemistry. 13: 7780-7789.

Oliver, J. E. and Whitfield, A. E. 2016. The genus Tospovirus: emerging bunyaviruses that threaten food security. Annual Review of Virology. 3: 101-124.

Pappu, H. R., Jones, R. A. C. and Jain, R. K. 2009. Global status tospovirus epidemics in diverse cropping systems: successes achieved and challenges ahead. Virus Research. 141: 219-236.

Pierschbacher, M. D. and Ruoslahti, E. 1984. Cell attachment activity of fibronectin can be duplicated by small synthetic fragments of the molecule. 309: 30-33.

Pierschbacher, M. and Ruoslahti, E. 1987. Influence of stereochemistry of the sequence Arg- Gly-Asp-Xaa on binding specificity in cell adhesion. The Journal Biological Chemistry. 262: 17294 - 17298.

Plow, E. F., Pierschbacher, M. D., Ruoslahti, E., Marguerie, G. and Ginsberg, M. H. 1987. Arginylglycyl-aspartic acid sequences and fibrinogen binding to platelets. Blood. 70: 110 - 115.

Pradel, G., Hayton, K., Aravind, L., Iyer, L. M., Abrahamsen, M. S., Bonawitz, A., Mejia, C., Templeton, T. J. 2004. A multidomain adhesion protein family expressed in Plasmodium falciparum is essential for transmission to the mosquito. The Journal of Experimental Medicine. 199: 1533-1544.

Riley, D. G, Joseph, S. V., Srinivasan, R. and Stanley, D. 2011. Thrips vectors of tospoviruses. Journal of Integrated Pest Management. 1: 1-10.

Rotenberg, D., Jacobson, A., Schneweis, D. J. and Whitfield, A. E. 2015. Thrips transmission of tospoviruses. Current Opinion in Virology. 15: 80-89.

Rudolph, C., Schreier, P. H. and Uhrig, J. F. 2003. Peptide-mediated broad-spectrum plant resistance to tospoviruses. Proceedings of the National Academy of Sciences of the United States of America. 100: 4429-4434.

Santos, M. J. D., Wigdorovitz, A., Trono, K., Ríos, R. D., Franzone, P. M., Gil, F., Moreno, J., Carrillo, C., Escribano, J. M. Borca, M. V. 2002. A novel methodology to develop a foot and mouth disease virus (FMDV) peptide-based vaccine in transgenic plants. Vaccine. 20: 7-8.

Schwab, E. H., Halbig, M., Glenske, K., Wagner, A-S., Wenisch, S., Cavalcanti-Adam, E. A. 2013. Distinct effects of RGD-glycoproteins on integrin-mediated adhesion and osteogenic

differentiation of human mesenchymal stem cells. International Journal of Medical Sciences. 10: 1846-1859.

Spizz, G. and Blackshear, P. J. 2001. Overexpression of the myristoylated alanine-rich C-kinase substrate inhibits cell adhesion to extracellular matrix components. Journal of Biological Chemistry. 276: 32264-32273. Ullman, D. E., Whitfield, A. E. and German, T. L. 2005. Thrips and tospoviruses come of age: mapping determinants of insect transmission. Proceedings of the National Academy of Sciences of the United States of America. 102: 4931-4932.

Wei, Y., Zhang, Y., Cai, H., Mirza, A. M., Iorio, R. M., Peeples, M. E. Niewiesk, S. and Li, J. 2014. Roles of putative integrin-binding motif of the human metapneumovirus fusion (f) protein in cell-cell fusion, viral infectivity, and pathogenesis. Journal of Virology. 88: 4338-4352.

Whitfield, A. E. 2004. Ph.D dissertation. Tomato spotted wilt virus acquisition by thrips: the role of the viral glycoproteins. Univerisity of Wisconsin-Madison, Madison, U. S.

Whitfield, A. E., Ullman, D. E., German, T.L. 2004. Expression and characterization of a soluble form of Tomato spotted wilt virus glycoprotein GN. Journal of Virology. 78: 13197-13205.

Whitfield, A. E., Ullman, D. E. and German, T. L. 2005. Tospovirus-thrips interaction. Annual review of phytopathology. 43: 459-489.

Whitfield, A. E., Kuman, N. K. K., Rotenberg, D., Ullman, D. E., Wyman, E. A., Zietlow, C., Willis, D. K. and German, T. L. 2008. A soluble form of the Tomato spotted wilt virus (TSWV) glycoprotein GN (GN -S) inhibits transmission of TSWV by Frankliniella occidentalis. Phytopathology. 98: 45-50.

Whitfield, A. E., Falk, B. W. and Rotenberg, D. 2015. Insect vector-mediated transmission of plant viruses. Virology. 479-480: 278-289.

Wild, C., Oas, T., McDanal, C., Bolognesi, D. and Matthews, T. 1992. A synthetic peptide inhibitor of human immunodeficiency virus replication: correlation between solution structure and viral inhibition. Proceedings of the National Academy of Sciences of the United States of America. 89: 10537-10541.

Woods, A., Couchman, J. R., Johansson, S., Hook, M. 1986. Adhesion and cytoskeletal organization of fibroblasts in response to fibronectin fragments. The EMBO Journal. 5: 665e70.

Yang, M-Y, Sunderland, K. and Mao, C-B. 2017. Virus-derived peptides for clinical applications. Chemical. Reviews. 117: 10377-10402.

Yu, Y-P., Wang, Q., Liu, Y-C and Xie, Y. 2014. Molecular basis for the targeted binding of RGD-contain peptide to integrin αVβ3. Biomaterials. 35: 1667-1675.

Zapp, C., Minsky, B. B. and Boehm, H. 2018. Tuning RGD motif and hyaluronan density to study integrin binding. Frontiers in Physiology. 9: 1022.

Zhang, Z., Pan, L., Ding, Y., Zhou, P., Lv, J. Chen, H., Fang, Y., Liu, X., Chang, H., Zhang, J., Shao, J., Lin, T., Zhao, F., Zhang, Y. and Wang, Y. 2015. Efficacy of synthetic peptide candidate vaccines against serotype-A food-and-mouth disease virus in cattle. Applied Microbiology Biotechnology. 99: 1389-1398.

Zhou, J., Kantartzi, S. K., Wen, R.-H., Newman, M., Hajimorad, M. R., Rupe, J. C., Tzanetakis, I. E. 2011. Molecular characterization of a new tospovirus infecting soybean. Virus Genes. 43: 289-295.

Zhou, J. and Tzanetakis, I. E. 2013. Epidemiology of Soybean vein necrosis-associated virus. Phytopathology. 103: 966-971.

Chapter V

Screening of Selected Soybean Accessions for Tolerance to the Soybean Thrips, Neohydatothrips variabilis (Beach)

Abstract

Soybean vein necrosis virus (SVNV), the causal agent of soybean vein necrosis disease, is primarily transmitted by Neohydatothrips variabilis (Beach) through feeding on permissive hosts. The widespread presence of the virus in soybean and its negative impact on seed oil content demonstrate the need for effective control strategies for the disease. Using germplasm with resistance to pathogens is the major component of viral disease management practice.

However, no SVNV-resistant soybean variety has been identified to date. Chemical control of thrips proved to be difficult because of physiological and behavioral characteristics of thrips.

Other than resistance to the pathogen, varieties that have resistance to the vector could modify vector behavior and reduce the incidence of transmission and disease. Such resistance may result from physical or biochemical features of particular varieties or the combination of the two.

Trichomes on plant surfaces are the first plant barrier against insect feeding and their density levels have been considered to be the basis of plant resistance to insect infestation. In this study, we evaluated feeding preference of N. variabilis on selected soybean accessions with differential leaf pubescence levels. We found feeding damage caused by thrips differs among accessions and is correlated to their pubescence levels.

Introduction

Plant morphological, structural and physiological characteristics including surface color,

waxiness, leaf thickness, pubescence and arrangement of vascular bundles influence insect

behavior and dynamics (Espelie et al., 1991; Csizinszky et al., 1995; Peter et al., 1995; Cohen et

al., 1996; Bernays 1998; Stafford et al., 2012) and consequently affect plant resistance acting

against herbivory feeders. Pubescence or trichome, is a hair-like appendage extending from the

plant epidermis of aerial tissues (Levis, 1973). Trichomes on leaf surface are the first plant

structures that insects contact with during the initial stages of host acceptance. Trichomes vary in

forms and functions; features including density, length and orientation could influence insect

feeding behaviors. The feeding preference of phytophagous insects are greatly affected by

trichome density as it alters the optical properties of the leaf surface and appears to be

cumbersome for insects to attach and move on leaf surface (Levin, 1973; Handley et al, 2005).

The variation of pubescence density is a major factor in determining plant resistance to insect

infestations and has frequently been the basis for breeding crops with resistance to insects

(Southwood, 1986; Lam and Pedigo, 2001).

The influence of soybean leaf pubescence on feeding and development preference has been investigated for several major insect groups including leafhoppers, aphids, beetles and whiteflies (Poos and Smith, 1931; Gunasinghe et al., 1988; Lambert, et al., 1995; Gannon and

Bach, 1996). In comparison, the effect of trichome density on the dynamics of thrips species feeding on soybean has not been studied in depth as thrips are generally considered minor pests, causing limited damage. Neohydatothrips variabilis is the primary vector of soybean vein necrosis virus (SVNV), the causal agent of the most prevalent viral disease in North America.

The virus affects soybean seed quality, especially oil content and has raisen the profile of

soybean-feeding thrips as virus vectors (Zhou and Tzanetakis, 2013). For this reason a thorough understanding of their feeding behavior is a prerequisite for virus control and disease management.

N. variabilis shares several physiological and behavioral characteristics with well-studied thrips species. Once established on host plants, they tend to hide deeply into unexpanded leaves and flowers where insecticides could not reach. Their high fecundity and protected egg and pupal stages also prevents an adequate spray coverage of potentially effective chemicals. Moreover, insecticides rarely kill viruliferous thrips in a timely manner to prevent further dissemination of plant viruses. To compensate the ineffectiveness of pesticides in thrips management, extensive efforts have been made to incorporate host resistance into breeding, an important component of integrated pest management (Coudriet et al., 1974; Zeier and Wright, 1995; Jensen, 2000; López et al., 2011; Maharijaya et al., 2011). However, similar work has not been conducted on the N. variabilis/soybean system. The objective of this study was to evaluate the feeding preference of

N. variabilis in relation to trichome density aiming to identify genetic resources that could be utilized in breeding thrips-resistant cultivars to reduce SVNV dispersal.

100

Materials and Methods

Plant materials

Eleven accessions with varying pubescence levels were selected by the soybean breeding

program at the University of Arkansas (Fayetteville, U.S.) and provided by USDA Soybean

Germplasm Collection (Urbana, U.S.) (Table 1).

Neohydatothrips variabilis colony and virus inoculation

A mini-aspirator was used to collect adults of N. variabilis from the upper canopy of field

grown soybean plants. The insects were released to soybean seedlings grown in 20” by 10”

planting trays and kept in growth chambers under controlled environment (27°C, Light: Dark =

16h: 8h). Trays were replaced with new seedlings every three weeks. Adults and larvae from old

trays were pooled and placed onto new seedling trays for thrips propagation. SVNV was

maintained within growth chambers as viruliferous thrips propagated and plants were tested

periodically using dot blot immunoassay and RT-PCR to confirm the presence of the virus as

previously described (Zhou and Tzanetakis, 2013; Zhou et al., 2018). N. variabilis adults (150-

200) were pooled on virus-free soybean leaf dishes to rear larvae under controlled environment

(22°C, Light: Dark = 16h: 8h). Larvae hatched from leaf dishes within a 24 h period were

transferred onto SVNV-infected tissue using a size 0 painting brush and allowed to feed for 48 h

for virus acquisition.

Screening for thrips – resistant/tolerant accessions

Initial screening

Accessions were evaluated in sets of twelve including ‘Hutcheson’ as the standard control. In each experiment, 30-35 individual plants of each accession were germinated in

101

planting trays and used for screening. All accessions were placed in the same growth chamber

under controlled conditions (25 C, Light: Dark = 16h: 8h). Seedlings at the unifoliate stage were

infested with early second instar larvae (4-5 days post hatching) that were fed on SVNV-free or

SVNV-infected tissue for 48 h during their first instar larvae stage. A cohort of ten larvae were transferred onto the leaf blades of each plant using size 0 camel painting brush. Three weeks later, all plants were rated for the feeding damage appeared on leaflets with the experiment being repeated twice.

Confirmatory screening

Based on the results of the initial screening, accessions that had lower feeding damage levels were further evaluated. A random complete block design was employed in the confirmatory screening experiment. Briefly, two blocks were set up in two growth chambers under the same controlled conditions (25°C, Light: Dark = 16h: 8h). In each block, three replicates were included. Each replicate consisted of a single plant of every selected accession along with one

‘Hutcheson’ plant; plants belonging to one replicate were grown in a one-gallon pot. At unifoliate stage, a cohort of second instar larvae (20) that was exposed to SVNV for 48 h immediately after their hatching from SVNV-free leaf dishes were placed on each individual plant. Three weeks post-inoculation, unifoliates and trifoliates of individual plant were collected and the damage caused by thrips feeding was measured.

Data analysis

Measurement of thrips feeding damage

The feeding scars on leaflets were processed using ImageJ2 according to the manufacturer’s instruction. For each leaflet, the area of whole leaf and the area without feeding

102

scars were measured by the software and the percentage of feeding damage was generated as

below:

Thrips feeding area = Leaf area – scar-free area

Percentage of feeding damage = Thrips feeding area/ leaf area × 100%

Statistical analysis

Percentage of feeding damage was analyzed using JMP Pro 13 (SAS Institute Inc., Cary, NC).

Significant effects of different treatments were determined (P ≤ 0.05). Least square means of accessions at unifoliate stage, trifoliate stage and the average of the two stages were separated by

Dunnett’s test, respectively.

Results

The initial screening did not show any difference in thrips feeding habits based in whether tissue was infected or not with SVNV (data not shown). For the reason only SVNV- infected material was used in the confirmatory experiments using PIs 547551, 548241, 547422 and 547467 that had less feeding damage compared with the other accessions based on visual observation. Severe and mild feeding damage were shown in Fig 1. For each plant, the percentage of feeding damage at unifoliate stage was measured on two unifoliate leaflets and their average was recorded; similarly, the percentage of feeding damage at trifoliate stage was measured on three trifoliate leaflets and their average was recorded in Table 2.

Comparison of the mean of the average feeding percentage of unifoliate and trifoliate stages revealed a significant difference among the five genotypes (P = 0.0091). PI 547467 showed significantly higher feeding damage compared to ‘Hutcheson’ (P = 0.018; Table 3), whereas for the rest there was no statistically significant difference with PI 547422 showing the

103

lowest amount of feeding damage (Fig 2). This result is in agreement with the analysis

performed on the unifoliate stage alone: the degree of thrips damage varied significantly (P =

0.006) among accessions and thrips feeding caused significantly more damage on PI 547467 (P =

0.005; Table 4) than ‘Hutcheson’. Feeding percentages on trifoliate leaflets alone did not show

pronounced difference (P = 0.2598) among accessions. Independent of leaf stage, PI 547422

constantly had the lowest feeding damage among all tested accessions (Fig. 3 and 4).

Additionally, there is weak correlation between feeding damage of unifoliate and trifoliate stages

(R = 0.367).

Discussion

Virus transmission by insects is a highly specific process which involves the interplay among

virus, vector and host (Ng and Falk, 2006; Gómez et al., 2009). Genetic resistance to viruses is a

commonly used strategy to control plant viral diseases (Kang et al., 2005; Gururani et al., 2012).

Resistance against vectors is based on the fact that material has either physical or biochemical

barriers that makes it less favorable for vector feeding and proliferation (Gunasinghe et al., 1988;

Gómez et al., 2009). Soybean varieties differ on the level of leaf pubescence or leaf trichome

density and have been grouped into glabrous, normal/sparse and densely pubescent accordingly

(Lambert and Kilen, 1989). The role of this physical characteristic plays in soybean’s resistance

against some of major soybean pests including beetles, aphids and whiteflies has been

extensively studied (Gunasinghe et al., 1988; Lambert et al, 1995; Gannon and Bach, 1996);

however, its influence on feeding preference of Neohydatothrips variabilis, the primary vector of soybean vein necrosis virus (SVNV) was unknown. In this study, we selected soybean accessions with varying levels of leaf pubescence and analyzed feeding damage caused by N. variabilis. We found that the feeding preference correlates with trichome density levels where

104

accessions with medium to high level of pubescence are less preferred by thrips compared with glabrous genotypes; on the other hand, all tested accessions exhibited typical SVNV symptoms suggesting none of them was resistant to the virus.

Initial screening of twelve accessions (including ‘Hutcheson’ as control) under the same insect pressure revealed differential feeding damage levels: all glabrous genotypes (547737,

547412, 547156) and three out of four sparse accessions (91160, 547721, 547468) exhibited severe feeding damage such as distorted and scarred appearances. Only one accession with dense pubescence level – 639683 sustained damage at comparable level to the above mentioned accessions. In comparison, thrips feeding only caused stippling on three dense pubescent genotypes (547551, 548241 and 547467) and one sparse genotype (547422). This result suggests that the presence or absence of trichomes may affect thrips feeding preference. The feeding damage on accessions with lower damage levels (PIs 547551, 548241, 547422 and 547467) was further evaluated in order to search for genotypes with thrips-resistant or tolerant characteristics which can be used in integrated pest management of SVNV or breeding of disease-resistant varieties. Soybean is the preferred host of N. variabilis in the presence of other plant species

(Keough et al., 2016; Irizarry et al., 2018) which suggests that N. variabilis could use almost any soybean variety as food source. Whether or not the soybean accession(s) thrips choose to feed on truly mirrors their feeding preference is largely dependent on the availability of their preferred genotypes. While evaluating thrips feeding preference among accessions, it is essential to eliminate any constraint of thrips movement as was done for PIs 547422, 547551, 547467 and

548241.

Development of standardization of screening techniques is the key for an effective evaluation. In this study, only seedlings at unifoliate stage were utilized as this is a very

105

susceptible host life stage (Zhou, personal observation). The use of seedlings at the same stage

also eliminated possible effects of leaf age. Individual plants were subjected to a uniform insect

pressure consisting of 20 second instar larvae minimizing differential feeding behaviors among stages. Another critical component affecting the effectivity of the evaluation system is the rating method. Probing and feeding of thrips on leaf epidermal and mesophyll cells cause stippling and scarred appearance on leaf surface when a large number of feeding wounds occurred close to each other. Wounds distributed sporadically on leaflets are usually separated from the striplings and scars and exhibit as small white spots, sometimes with a black center in the middle. These scattered micro-wounds are difficult to measure in a symptom rating scale based solely on observation. To tackle this issue, we applied an image processing program – ImageJ2 (Rueden et

al., 2017) to calculate feeding percentage for each leaflet collected.

Glabrous accessions are most susceptible to feeding damages caused by N. variabilis in

comparison with accessions carrying trichomes (Fig 1). This results are consistent with findings

of previous studies where different species of arthropods including the potato leafhopper, beet

armyworm and agromyzid beanflies caused significantly higher feeding damage on glabrous

soybean isolines than the pubescent ones (Chiang and Norris, 1983; Elden and Lambert, 1992;

Tillman et al., 1997). It was also suggested that trichome density may shape the population

dynamics of Mexican bean beetle (Epilachna varivestris) and densely pubescent soybean

cultivars had the potential to inhibit bean leaf beetle feeding on pods (Gannon and Bach., 1996).

Possible explanations of this phenomenon relates to the proposed physiological functions of dense pubescence such as reflecting or absorbing light at certain wavelengths (Peter al., 1995). In the absence or relatively sparse presence of trichomes, glabrous accessions might be visually

more attractive to thrips given that the true color of leaflets which is preferred by thrips are not

106

masked by surface hairs. Apart from that, the lack of hairs is also a stimulus for thrips movement and selection of preferred feeding spot on leaf blades.

In the confirmatory screening, PI 547422 - a sparely pubescent accession, exhibited lowest feeding damage when compared with 547551, 548241, 547467, all of which are categorized as densely pubescent accessions. This result does not agree with our hypothesis that thrips feeding damage decreases with increasing trichome density. It is possible that other than functioning as purely structural barrier for insect behaviors, trichomes also conserve heat and moisture by trapping a layer of air against the leaf surface (Peter al., 1995). Following this line of logic, dense pubescence could serve as optimal oviposition substrate for thrips and thereby attracting them to feed, although dense trichomes do create irregular surface that is less favorable for the locomotion of thrips individuals. It is also likely the orientation and/or length of trichomes instead of density influence thrips feeding behaviors. Soybean leaves possess simple non-glandular trichomes with the length of 1-3 mm, either straight or curly (Peter al., 1995).

Morphologically, some surface hairs are perpendicular to leaf surface; some are lying in close proximity to surface, whereas others are tilted with certain angels (Turnipseed, 1977; Lambert et al., 1995). When analyzing the best correlation between trichome variation and whitefly populations in soybean, host genotypes with trichomes lying flat against the leaf surface had fewer whitefly infestations in comparison with their erect counterparts (Lambert et al., 1995). In addition, size of the insect body also affects which parameter of trichome determines the degree of resistance. Turnispeed (1977) found that the population of potato leaf hopper (Empoasca fabae, body length = 1.0-4.0 mm) decreases with increasing soybean trichome length, regardless of trichome density; whereas for springtail (Deuterosmiathurus yumanensis, body length = 0.2-

0.4 mm) population decreased with an increasing trichome density. N. variabilis has the body

107

length of 1.0-1.5 mm, which falls into the first category of Turnispeed’s study. It is possible

trichome length rather than its density is the prime factor in soybean resistance to thrips. Other

than physical causes for host resistance to insects, studies on different Fabaceae species also

revealed varietal resistance as a result of biochemical factors (Beck, 1965). Phaseolus varieties

containing high levels of glycosides is resistant to the Mexican bean beetle whereas the elevated

level of trypsin, which differentiates resistant from non-resistant Vigna unguiculata varieties, is

inhibitory for bruchid beetle (Callosobruchus maculatus) damage (Nayar and Fraenkel, 1963;

Gatehouse, et al., 1979). The resistance of PI 547422 may also result from its possession of

biochemical characteristics having adverse effects on thrips feeding activities such as lower

content of sugars and amino acids (Beck, 1965). The genetic variability of sugar accumulation in

soybean leaves (Zhao et al., 2008) suggests germplasm may vary in the ability of maintaining

feeding activities once initial probing on leaf surface occurs.

Feeding damage caused by N. variabilis on soybean leaflets reflects two independent events including 1) the feeding preference of thrips and 2) host plant resistance for thrips feeding. These events entail not only physical barriers such as trichomes and epicuticular wax on plant surface but also a complex series biochemical elements including the accumulation of high level of secondary metabolites (Sadasivam and Thayumanavan, 2003; Singer et al., 2003). The feeding preference of N. variabilis on selected material could be the combination of physical and biochemical defense mechanisms. On the other hand, the expression of plant resistance to insects is not only dependent on host genotypes but also in relation to environmental conditions including photoperiod, temperature, water supply and soil nutrients, factors that could alter the production and accumulation of secondary metabolites of plant (Gershenzon, 1984). Another important factor that could contribute to the results is insect density. The 20 thrips transferred to

108

individual plant in this study was sufficient to cause visible feeding damage and high virus transmission rate (Zhou and Tzanetakis, 2013). It should be noted that in addition to feeding preference, other N. variabilis behaviors on soybean including reproductive capacity, longevity, mortality, and population dynamics need to be investigated in order to determine their preferences on different accessions.

109

Table 1. Selected soybean accessions and their pubescence type.

Accession Name Pubescence Type

PI 91160 Sparse PI 547721 Sparse PI 547468 Sparse PI 547422 Sparse PI 639683 Dense PI 547551 Dense PI 547467 Dense PI 548241 Dense PI 547737 Glabrous PI 547412 Glabrous PI 594156 Glabrous

110

Fig 1. Feeding damage caused by N. variabilis during the initial screening. First panel represents severe damage (first row: PIs 547737, 547412, 594156 and 91160; second row: PIs 547468, 547721 and 639683); second panel represents mild damage (PIs 548241, 547422, 547467 and 547551).

111

Table 2. Percentage of thrips feeding damage on accessions 547551, 548241, 547422 and 547467

Feeding Percentage (%) Accession Block No. of Plant Name Unifoliate Trifoliate Average

1 12.09 5.49 8.79 PI 547551 2 7.34 6.30 6.82 3 3.56 9.28 6.42 1 8.81 7.70 8.26 PI 548241 2 8.33 6.89 7.61 3 3.43 11.40 7.42 1 3.03 7.05 5.04

1 PI 547422 2 3.36 4.06 3.71

3 3.32 4.48 3.90

1 9.76 4.62 7.19 PI 547467 2 10.26 11.85 11.06 3 9.82 6.06 7.94 1 4.12 7.61 5.87 Hutcheson 2 4.66 3.73 4.20 3 3.97 5.56 4.77 4 13.19 12.27 12.73 PI 547551 5 5.23 5.38 5.31

6 4.52 5.82 5.17

4 15.56 11.22 13.39

PI 548241 5 2.31 5.43 3.87 6 6.44 8.89 7.67 2 4 3.47 5.74 4.61 PI 547422 5 2.63 3.08 2.86 6 2.95 4.14 3.55 4 6.35 6.97 6.66 PI 547467 5 11.04 5.83 8.44 6 17.26 6.14 11.70 4 3.32 4.32 3.82 Hutcheson 5 3.82 2.98 3.40 6 3.74 10.29 7.02

112

Table 3. Comparison of least squares means (± SE) of the average of unifoliate and trifoliate stages for each soybean accessions using Dunnett’s method.

Accession LSM ± SE P-value

Hutcheson 4.9 ± 0.92 - PI547422 3.9 ± 0.92 0.892

PI547467 8.8 ± 0.92 0.018 PI547551 7.5 ± 0.92 0.148

PI548241 8.0 ± 0.92 0.070

Fig 2. Feeding percentage of the average of unifoliate and trifoliate stages for each accession.

113

Table 4. Comparison of least squares means (± SE) of soybean accessions at unifoliate stage using Dunnett’s method.

Accession LSM ± SE P-value

Hutcheson 3.9 ± 1.31 -

PI547422 3.1 ± 1.31 0.977

PI547467 10.8 ± 1.31 0.005 PI547551 7.7 ± 1.31 0.171

PI548241 7.5 ± 1.31 0.202

Fig 3. Feeding percentage of different accessions at unifoliate stage.

114

Table 5. Comparison of least squares means (± SE) of accessions at trifoliate stage using Dunnett’s method.

Accession LSM ± SE P-value

Hutcheson 5.8 ± 1.03 -

PI547422 4.8 ± 1.03 0.901 PI547467 6.9 ± 1.03 0.840

PI547551 7.4 ± 1.03 0.614

PI548241 8.6 ± 1.03 0.189

Fig 4. Feeding percentage of different accessions at trifoliate stage.

115

References

Beck, S. D. 1965. Resistance of plants to insects. Annual Review of Entomology. 10: 207-232.

Bernays, E. A. 1998. Evolution of feeding behavior in insect herbivores. Bioscience. 48: 35-44.

Chiang, H. S., and Norris, D. M. 1983. Morphological and physiological parameters of soybean resistance to agromyzid beanflies. Environmental Entomology. 12: 260 - 265.

Cohen, A. C., Henneberry, T. J. and Chu, C. C. 1996. Geometric relationships between whitefly feeding behavior and vascular bundle arrangements. Entomologia Experimentalis et Applicata. 78: 135-142.

Coudriet, D. L., Kishaba, A. N., McCreight, J. D. and Bohn, G. W. 1974. Varietal resistance in onions to thrips. Journal of Economic Entomology. 72: 614-615.

Csizinszky, A. A., Schuster, D. J. and Kring, J. B. 1995. Color mulches influence yield and insect pest populations in tomatoes. Journal of the American Society for Horticultural Science. 120: 778-784.

Elden, T. C., and Lambert, L. 1992. Mechanisms of potato leafhopper resistance in soybean lines isogenic for pubescence type. Crop Science. 32: 1187 -1191.

Espelie, K. E., Bernays, E. A. and Brown, J. J. 1991. Plant and insect cuticular lipids serve as behavioral cues for insects. Insect Biochemistry and Physiology. 17: 223-233.

Gannon, A. J. and Bach, C. E. 1996. Effects of soybean trichome density of Mexican bean beetle (Coleoptera: Coccinellidae) development and feeding preference. Environmental Entomology. 25: 1077-1082.

Gatehouse, A. M. R., Gatehouse, J. A., Dobie, P., Kilminster, A. M. and Boulter, D. 1979. Biochemical basis of insect resistance in Vigna unguiculata. Journal of the Science of Food and Agriculture. 30: 948-958.

Gershenzon, J. 1984. Changes in the levels of plant secondary metabolites under water and nutrient stress. pp 273-320 in Phytochemical Adaptations to Stress. Timmermann, B. N., Steelink, C. and Loewus, F. A. (eds). Plenum Press. New York. U. S.

Gómez, P., Rodríguez-Hernández, A. M., Moury, B. and Aranda, M. A. 2009. Genetic resistance for the sustainable control of plant virus disease: breeding, mechanisms and durability. European Journal of Plant Pathology. 125: 1-22.

Gunasinghe, U. B., Irwin, M. E. and Kampmeier, G. E. 1988. Soybean leaf pubescence affects aphid vector transmission and field spread soybean mosaic virus. Annals of Applied Biology. 112: 259-272.

116

Gururani, M. A., Venkatesh, J., Upadhyaya, C. P., Nookaraju, A., Pandey, S. K. and Park, S. W. 2012. Plant disease resistance genes: current status and future directions. Physiological and Molecular Plant Pathology. 78: 51-65.

Handley, R., Ekbom, B and Ågren, J. 2005. Variation in trichome density and resistance against a specialist insect herbivore in natural populations of Arabidopsis thaliana. Ecological Entomology. 30: 284-292.

Jensen, S. E. 2000. Insecticide resistance in the western flower thrips, Frankliniella occidentalis. Integrated Pest Management Reviews. 5: 131-146.

Kang, B. C., Yeam, I. and Jahn, M. M. 2005. Genetics of plant virus resistance. Annual Review of Phytopathology. 43: 581-621.

Keough, S., Han, J., Shuman, T., Wise, K. and Nachappa, P. 2016. Effects of Soybean vein necrosis virus on life history and host preference of its vector, Neohydatothrips variabilis, and evaluation of vector status of Frankliniella tritici and Frankliniella fusca. Journal of Economic Entomology. 109: 1979-1987.

Lam, W. F. and Pedigo, L. P. 2001. Effect of trichome density on soybean pod feeding by adult bean leaf beetles (Coleoptera: Chrysomelidae). Journal of Economic Entomology. 94: 1459- 1463.

Lambert, L. and Kilen, T. C. 1989. Influence and performance of soybean lines isogenic for pubescence type on oviposition preference and egg distribution of corn earworm (Lepidoptera: Noctuidae). Journal of Entomological Science. 24: 309-316.

Lambert, A. L., McPherson, R. M. and Espelie, K. E. 1995. Soybean host plant resistance mechanisms that alter abundance of whiteflies (Homoptera: Aleyrodidae). Environmental Entomology. 26: 1381-1386.

Levin, D. A. 1973. The role of trichomes in plant defense. The Quarterly Review of Biology. 48: 3-15.

López, C., Aramburu, J., Galipienso, L., Soler, S., Nuez, F. and Rubio, L. 2011. Evolutionary analysis of tomato Sw-5 resistance-breaking isolates of Tomato spotted wilt virus. Journal of General Virology. 92: 210-215.

Irizarry, M. D., Elmore, M. G., Batzer, J. C., Whitham, S. A. and Mueller, D. S. 2018. Alternative hosts of Soybean vein necrosis virus and feeding preferences of its vector soybean thrips. Plant Health Progress. 19: 176-181.

Maharijaya, A., Vosman, B., Steenhuis-Broers, G., Harpenas, A., Purwito, A. and Visser, R. G. F. 2011. Screening of pepper accessions for resistance against two thrips species (Frankliniella occidentalis and Thrips parvispinus). Euphytica. 177: 401-410.

117

Nayar, J. K. and Fraenkel, G. 1963. The chemical basis of host selection in the Mexican bean beetle, Epilachna varivestris. Annals of the Entomological Society of America. 56: 174-178.

Ng, J. C. and Falk, B. W. 2006. Virus-vector interactions mediating nonpersistent and semipersistent transmission of plant viruses. Annual Review of Phytopathology. 44: 183-212.

Peter, A. J., Shanower, T. G. and Romeis, J. 1995. The role of plant trichomes in insect resistance: a selective review. Phytophaga. 7: 41-63.

Poos, F. W. and Smith, F. F. 1931. A comparison of oviposition and nymphal development of Empoasca fabae Harris on different host plant. Journal Economic Entomology. 84: 361-371.

Rueden, C. T., Schindelin, J. and Hiner, M. C., DeZonia, B. E., Walter, A. E., Arena, E. T. and Eliceiri, K. W. 2017. ImageJ2: ImageJ for the next generation of scientific image data. BMC Bioinformatics. 18: 529.

Sadasivam, S. and Thayumanavan, B. B. 2003. Molecular host plant resistance to pests. pp. 479. CRC Press. New York. U. S.

Singer, A. C., Crowley, D. E., and Thompson, I. P. 2003. Secondary plant metabolites in phytoremediation and biotransformation. Trends in Biotechnology. 21: 123 - 130.

Southwood, R. 1986. Plant surfaces and insects - an overview, pp. 1-22. In B. Juniper and R. Southwood (eds.), Insects and the plant surface. Edward Arnold, Baltimore, MD, U. S.

Stafford, C. A., Gregory, P. W. and Ullman, D. E. 2012. Hitching a ride: vector feeding and virus transmission. Communicative and Integrative Biology. 5: 43-49.

Tillman, P. G., Lambert, L., and Mulrooney, J. E. 1997. Influence of soybean pubescence on abundance of beet armyworm, Spodoptera exigua. Southwest Entomologist. 22: 459 - 460.

Turnipseed, S. G. 1977. Influence of trichome variations on populations of small phytophagous insects in soybean. Environmental Entomology. 6: 815-817.

Zeier, P. and Wright, M. G. 1995. Thrips resistance in Galdiulus spp.: potential for IPM and breeding. Thrips Biology and Management. 276: 411-416.

Zhao, L-M., Zheng, D-F., Feng, N-J., Dum J-D., Xiang, H-T., Zhang, Y-X. 2008. Effects of plant growth regulators (PGRs) on photosynthetic characteristics and sugar accumulation in soybean leaves. 2008-03.

Zhou, J. and Tzanetakis, I. E. 2013. Epidemiology of Soybean vein necrosis-associated virus. Phytopathology. 103: 966-971.

118

119

Conclusions

Soybean vein necrosis virus (SVNV), the causal agent of soybean vein necrosis disease

(SVN), has become the most prevalent virus infecting soybean in North America (Zhou et al.,

2011; Zhou, 2012; Ali and Abdalla, 2013; Conner et al., 2013; Escalante et al., 2018; Han et al.,

2013; Jacobs and Chilvers., 2013; Kleczewski, 2016; Smith et al., 2013). Because of that, it is imperative to further understand the virus epidemiology. None of the alternative hosts of SVNV identified in previous studies has the potential to act as a green-bridge in winter months (Zhou and Tzanetakis, 2013). Survey on indigenous weed species revealed that kudzu (Pueraria montana), present in millions of acres in Southeastern United States, is an asymptomatic, systemic host of the virus (Zhou et al., 2018). One of the major concerns for any newly characterized virus, including SVNV is its coinfection and synergisms with other viruses. Co- infections with bean pod mottle virus (BPMV) and soybean mosaic virus (SMV), the most economically important viruses infection soybean (Hartman, 2015) were evaluated and it was determined that SVNV could move systemically with the assistance of BPMV.

SVNV is transmitted very efficiently by Neohydatothrips variabilis (Beach), its primary vector in the field (Zhou and Tzanetakis, 2013; Keough et al., 2016). The ineffectiveness of pesticides in thrips management highlights the necessity of identifying resistance against N. variabilis. Genotypes that have resistance to the vector could modify vector behavior and reduce the incidence of transmission and disease. Trichomes on plant surfaces are the first physical barrier of plants against insect feeding and their density could be the basis of resistance to insects. We found that feeding preference of N. variabilis on soybean accessions is weakly correlated to the leaf pubescence levels.

120

Virus entry into host cells is the prerequisite of virus infection process and is mediated by the interaction of orthotospovirus-coded glycoproteins and cellular receptors (Whitfield et al.,

2005; Ullman et al., 2015). Several motifs, including RGD and N-linked glycosylation sites on the glycoproteins have been recognized as critical for cell entry of the virus (Whitfield, 2004;

Whitfield et al., 2004; Whitfield et al., 2008). The comparison of transmission efficiency of thrips fed on polypeptides containing target sequences indicated that blocking putative receptors prior to virus acquisition could significantly reduce the transmission efficiency of the virus.

121

References

Ali, A. and Abdalla, O. A. 2013. First report of Soybean vein necrosis virus in soybean fields of Oklahoma. Plant Disease. 97: 1664.

Han, J., Domier, L. L., Dorrance, A. E. and Qu, F. 2013. First report of Soybean vein necrosis- associated virus in Ohio soybean fields. Plant Disease. 97: 693.

Jacobs, J. L. and Chilvers, M.I. 2013. First report of Soybean vein necrosis virus on soybeans in Michigan. Plant Disease. 97: 1387.

Kleczewski, N. 2016. Research updates on soybean vein necrosis virus. http://extension.udel.edu/fieldcropdisease/2016/01/20/research-updates-on-soybean-vein- necrosis-virus

Ullman, D. E., Whitfield, A. E. and German, T. L. 2005. Thrips and tospoviruses come of age: mapping determinants of insect transmission. Proceedings of the National Academy of Sciences of the United States of America. 102: 4931-4932.

Whitfield, A. E. 2004. Ph.D dissertation. Tomato spotted wilt virus acquisition by thrips: the role of the viral glycoproteins. Univerisity of Wisconsin-Madison, Madison, U. S.

122

Whitfield, A. E., Ullman, D. E., German, T.L. 2004. Expression and characterization of a soluble form of Tomato spotted wilt virus glycoprotein GN. Journal of Virology. 78: 13197-13205.

Whitfield, A. E., Ullman, D. E. and German, T. L. 2005. Tospovirus-thrips interaction. Annual review of phytopathology. 43: 459-489.

Zhou, J., Kantartzi, S. K., Wen, R.-H., Newman, M., Hajimorad, M. R., Rupe, J. C., Tzanetakis, I. E. 2011. Molecular characterization of a new tospovirus infecting soybean. Virus Genes. 43: 289-295.

Zhou, J. 2012. Characterization and epidemiology of Soybean vein necrosis associated virus. Thesis, University of Arkansas, Fayetteville, AR.

Zhou, J. and Tzanetakis, I. E. 2013. Epidemiology of Soybean vein necrosis-associated virus. Phytopathology. 103: 966-971.

123