PHEROMONE-BASED MONITORING of Pseudococcus Maritimus and INTRA

PHEROMONE-BASED MONITORING OF Pseudococcus maritimus AND INTRA

AND INTERSPECIFIC VECTOR EFFICIENCY OF Parthenolecanium corni

AND Pseudococcus maritimus AMONG AND BETWEEN

Vitis x labruscana L. AND Vitis vinifera

BRIAN WILLIAM BAHDER

A dissertation submitted in partial fulfillment of the requirements for the degree of

DOCTOR OF PHILOSOPHY

WASHINGTON STATE UNIVERSITY Department of Entomology

MAY 2013

To the Faculty of Washington State University:

The members of the Committee appointed to examine the dissertation of BRIAN

WILLIAM BAHDER find it satisfactory and recommend that it be accepted.

Douglas B. Walsh, Ph.D., Chair

Naidu A. Rayapati, Ph.D.

David R. Horton, Ph.D.

ACKNOWLEDGEMENTS

I would like to thank Dr. Douglas B. Walsh for giving me this opportunity and

allowing me to attain the level of knowledge and training that I have obtained. I would

like to thank Dr. Naidu A. Rayapati for allowing me to work in his laboratory so that I

could expand my knowledge and base and receive invaluable training in a new field of

research. I thank my committee members Dr. Keith Pike and Dr. David Horton in their

critical review of my manuscript. I would like to thank Dr. Olufemi Alabi, Dr. Sridhar

Jarugula, and Sudarsana Poojari for teaching my all of the molecular techniques needed for achieving our objectives. I would also like to thank all of my friends in Delaware,

Florida, and Washington who provided support, both intellectual and personal. I would like to thank my family; mom, dad, Jaime, and Michelle for helping me get to where I am. Finally I would like to thank my wife, Luz Denia Bahder for all of her love and support, making this journey much smoother and much more valuable.

iii PHEROMONE-BASED MONITORING OF Pseudococcus maritimus AND INTRA

AND INTERSPECIFIC VECTOR EFFICIENCY OF Parthenolecanium corni

AND Pseudococcus maritimus AMONG AND BETWEEN

Vitis x labruscana L. AND Vitis vinifera

Abstract

By Brian William Bahder, Ph.D. Washington State University May 2013

Chair: Douglas B. Walsh

Grapes have a major economic impact on the state of Washington, which is the

number one producer of Concord juice grapes in the United States and the second largest

producer of wine grapes in the United States, after California. Due to the importance of

grapes to Washington State, there is a very low tolerance to pests and diseases. The most

devastating viral disease of grapes, grapevine leafroll disease, was found in Washington

State in 2005 and due to the presence of natural vectors Pseudococcus maritimus and

Parthenolecanium corni of the associated viruses, a more rapid means of detection and

better understanding of the epidemiology of the disease is necessary. The primary

objectives of this research are to develop a rapid, cost-effective monitoring program for

Ps. maritimus using its sex pheromone and to determine if Pa. corni and Ps. maritimus are capable of transmitting the viruses associated with grapevine leafroll disease between two different species of grape.

To establish an economical means to monitor for Ps. maritimus, three different trapping densities were deployed in various vineyards; one, four, and eight pheromone- baited traps per 12.14 hectares of vineyards. Traps were collected weekly in 2010 and

iv 2011 and adult males were counted. In this project we were able to demonstrate that one

trap per 12.14 ha of vineyard was adequate to detect populations of Ps. maritimus so as to make appropriate management decisions.

To determine if both vector species were capable of transmitting this virus between species, a total of 90 bioassays were conducted under greenhouse conditions where first instars of each species was allowed a one week access period on a virus infected plant and were then transferred to a virus free recipient plant for a one week inoculation period. With this research, we were able to demonstrate that both species of vector were capable of transmitting this virus between species of grape under greenhouse conditions and that Ps. maritimus appears to be a more efficient vector than Pa. corni.

These results will be important in further understanding the epidemiology of grapevine leafroll disease and how to monitor and control its vectors.

v TABLE OF CONTENTS

ACKNOWLEDGEMENTS ...... iii

ABSTRACT ...... iv

LIST OF TABLES ...... viii

LIST OF FIGURES ...... ix

DEDICATION ...... xii

CHAPTER

1. INTRODUCTION ...... 1

History and Economics of Wine and Grapes in Washington State ...... 1

Grapevine Leafroll Disease and Associated Viruses ...... 2

Vector and Pathogen Interactions ...... 5

Grapevine Leafroll Disease in Washington State ...... 7

Insect Vectors of Grapevine Leafroll Disease ...... 9

References ...... 14

2. PHEROMONE-BASED MONITORING OF PSEUDOCOCCUS MARITIMUS (HEMIPTERA: PSEUDOCOCCIDAE) POPULATIONS IN CONCORD GRAPE VINEYARDS ...... 21

Introduction ...... 21

Materials and Methods ...... 24

Field Test Sites ...... 24

Pheromone-Baited Traps ...... 24

Trap Density ...... 24

Data Collection and Analysis ...... 25

Sample Size ...... 26

vi TABLE OF CONTENTS

Results ...... 27

Seasonal Phenology ...... 27

Trap Density Comparisons ...... 28

Sample Size ...... 30

Mealybug species ...... 35

Discussion ...... 35

References ...... 42

3. PSEUDOCOCCUS MARITIMUS (HEMIPTERA: PSEUDOCOCCIDAE) AND PARTHENOLECANIUM CORNI (HEMIPTERA: COCCIDAE) ARE CAPABLE OF TRANSMITTING GRAPEVINE LEAFROLL-ASSOCIATED VIRUS 3 BETWEEN VITIS x LABRUSCANA L. AND VITIS VINIFERA ...... 50

Introduction ...... 50

Materials and Methods ...... 53

Insect Colonies and Plant Material ...... 53

Identification of Pa. corni and Ps. maritimus ...... 54

Transmission Bioassays ...... 56

Data Analysis ...... 57

Results ...... 58

Species Identity ...... 58

Transmission Bioassays ...... 58

Discussion ...... 60

References ...... 66

vii LIST OF TABLES

CHAPTER 1

1-1. Modes of Transmission of insect vectors with piercing-sucking mouthparts that transmit viruses ...... 6

1-2. All species of virus detected in Washington State vineyards to date ...... 8

1-3. Species of GLRaVs and their associated vectors ...... 11

CHAPTER 3

3-1. Primer sequences for virus and insect verification ...... 55

viii LIST OF FIGURES

CHAPTER 1

1-1. A map of Washington State showing wine grape-growing regions in different American Viticultural Areas (AVAs) ...... 2

1-2. Grapevine leafroll disease in red-fruited varieties (A), white-fruited varieties (B), Concord juice grapes, V. x labruscana (C), and its distribution within vineyards of V. vinifera (D) ...... 4

1-3. Genome organization of GLRaV-3 ...... 5

1-4. Vitis vinifera vineyard grown adjacent to a Vitis labruscana vineyard ...... 12

1-5. Adult female of Pa. corni (A), adult female of Ps. maritimus (B), and adult male of Ps. maritimus (C) ...... 13

CHAPTER 2

2-1. Mean numbers (±SE) of adult male Pseudococcus maritimus recorded per trap per week in (A) 2010 (23 sample dates) and (B) 2011 (28 sample dates) show a significant difference among 12.14 ha vineyard blocks sampled (ANOVA for 2010: F = 18.63, df = 6, 2885, P < 0.001; 2011: F = 16.32, df = 8, 1083, P < 0.001). The vineyards corresponding to the abbreviations shown on the x-axis are listed in the methods section ...... 29

2-2. Mean numbers (±SE) of adult male Pseudococcus maritimus recorded per trap per week in 2010 and 2011 show two distinct flight periods in each year ...... 30

2-3. Mean numbers (±SE) of adult male Pseudococcus maritimus recorded per trap per week for trapping densities of one, four, and eight traps per 12.14 ha in (A) 2010 and (B) 2011 ...... 31

2-4. Regression curve between the natural logarithm (ln) of the variance and the ln of the mean (x) number of adult male Pseudococcus maritimus captured to obtain the coefficients of Taylor’s power law for data collected in 2010 showing positive relationship between ln variance and ln mean when data were analyzed by (A) vineyard (y = 1.53x + 0.537; F = 3797, df = 1, 146; P < 0.001; r2 = 0.963)or (B) treatment (y = 1.783x + 0.444; F = 3087; df = 1, 65; P < 0.001; r2 0.979) ...... 32

ix LIST OF FIGURES

2-5. Relationship between optimal sample size (pheromone traps per 12.14 ha vineyard) and adult male Pseudococcus maritimus per trap per week for sample size models based on Taylor’s Power Law (eq. 1) where a = 1.71 and b = 1.53, Iwao’s regression parameters (eq. 2) where α = 2.16 and β = 1.19, and the K parameter of the negative binomial (eq. 3) where Kc = 5.97. Data were from 2010 trap captures, with mean – variance relationships determined from each 13 trap data set collected in each of seven sampled vineyards on 23 sample dates ...... 33

2-6. (A) Relationship between optimal sample size (pheromone traps per 12.14 ha vineyard) and adult male Pseudococcus maritimus per trap per week using the K parameter of the negative binomial (eq. 3) where Kc = 5.97 (all 2010 trap data, as in Fig. 5), Kc = 8.92 (pheromone traps with >50 males per trap) and Kc = 10.50 (pheromone traps with >100 males per trap) and Kc = 10.50; and (B) number of pheromone traps (samples) per 12.14 ha vineyard needed under different precision errors from 5 to 40% as determined using Kc (eq. 4) determined from all 2010 trap data, traps with > 50 males, and traps with >100 males per trap per sample date ...34

CHAPTER 3

3-1. Sequence data for the COI gene of Ps. maritimus; Pmar_COI_Daane represents sequence data for Ps. maritimus provided by Daane et al. 2011, Pmar_COI represents COI sequence data obtained from individuals used in transmission bioassays, and Pvib_COI represent COI sequence data from Ps. viburni, the sister species to Ps. maritimus ...... 59

3-2. Sequence data for the COI gene of Pa. corni; Pcor_COI_Deng represents sequence data for Pa. corni provided by Deng et al. (JQ795617.1) and Pcor_COI represents sequence data for Pa. corni used in transmission bioassays ...... 60

3-3. Sequence data for hsp-70 gene of GLRaV-3 used in transmission bioassays; sequence data obtained from virus in source plant (1), sequence data obtained from virus within the insect vector (2), and sequence data obtained from virus in recipient Concord plant (3) ...... 61

3-4. Gel from RT-PCR test of recipient V. x labruscana L. plants exposed to first instars of Ps. maritimus that had been exposed to the virus infected V. vinifera source; C = Concord and number represents plant number ...... 62

3-5. Gel from RT-PCR test of recipient V. x labruscana L. plants exposed to first instars of Pa. corni that had been exposed to the virus infected V. vinifera source; C = Concord and number represents plant number ...... 62

x LIST OF FIGURES

3-6. Gel from RT-PCR test of recipient V. x labruscana L. plants exposed to first instars of Ps. maritimus that had been exposed to the virus infected V. x labruscana L. source; C = Concord and number represents plant number ...... 62

3-7. Gel from RT-PCR test of recipient V. x labruscana L. plants exposed to first instars of Pa. corni that had been exposed to the virus infected V. x labruscana L. source; C = Concord and number represents plant number ...... 64

3-8. Gel from RT-PCR test of recipient V. vinifera plants exposed to first instars of Ps. maritimus that had been exposed to the virus infected V. x labruscana L. source; W = Wine grape and number represents plant number ...... 64

3-9. Gel from RT-PCR test of recipient V. vinifera plants exposed to first instars of Pa. corni that had been exposed to the virus infected V. x labruscana L. source; W = Wine grape and number represents plant number ...... 67

Dedication

This dissertation is dedicated to my wife, Luz Denia Bahder, who provided invaluable emotional support and helped me achieve my dream.

xii CHAPTER ONE

INTRODUCTION

History and Economics of Wine and Grapes in Washington State

Grapes are the most valuable horticultural crop in the world, comprising about

eight million hectares (ha) of vineyard (Myles et al. 2011). In Washington State, there

are about 17,806 ha of European wine grapes (Vitis vinifera cv.), contributing about $8.6

billion at the state level and $14.9 billion at the national level (Stonebridge Research

Report April 2012). There are about 10,117 ha of Concord Juice grapes (Vitis x labruscana L.) currently grown in Washington State (Ball et al. 2002). Prior to 1969, there was virtually no wine industry in Washington State. A series of laws made wine grape production impractical, allowing for only small, home production of wine from

Vitis vinifera cvs and from Vitis x labruscana L. In 1969, a series of legislative hearings were held in Yakima and Seattle to determine if the laws that were inhibiting the growth of the wine industry should be overturned. Two scientists from Washington State

University are directly responsible for overturning the legislature, Walter Clore and Chas

Nagel. Following the overturning of these laws, numerous wineries began emerging in

Washington in the 1970s and since, have expanded and made Washington State the one of the leading wine producers in the United States, second only to California. Currently, there are 11 American Viticultural Areas (AVAs) in the state of Washington. These

Walter regions are specific wine grape growing areas as defined by the Alcohol and

Tobacco Tax and Trade Bureau (Figure 1-1). One of the main advantages that

Figure 1-1. A map of Washington State showing wine grape-growing regions in different

American Viticultural Areas (AVAs).

(Source: Cascade Valley Wine Country, 2012)

Clore saw to growing V. vinifera in Washington over California is the lack of many pests and diseases. Much of the planting stock for Washington vineyards is produced in

California (Martin et al. 2005). Despite having fewer pests, disease, and poor conditions for them to establish, the transportation of V. vinifera from California undoubtedly brought some pathogens along with them, the most problematic being grapevine leafroll disease (GLRD).

Grapevine Leafroll Disease and Associated Viruses

Grapevine leafroll disease is the most significant viral disease of V. vinifera in the world (Walker et al. 2004) and is responsible for about 60% of yield losses annually in

2 grape production worldwide (Naidu et al. 2008). Grapevine leafroll disease was first

described in the 19th century in Europe (Fuchs 2007) and symptoms of this disease in the

United States were first reported in 1946 in California (Harmon and Snyder 1946).

Symptoms vary greatly depending on cultivar and environmental conditions, however, in

all affected varieties of grape, there is a downward rolling of the leaf margins. In red-

fruited grape varieties there is a reddening of the intervenal region of the leaf while in

white-fruited varieties, there are varying degrees of chlorosis in the intervenal regions of

the leaves (Figure 1-2). In the ‘Concord’ variety of Vitis x labruscana L., GLRD does

not appear to produce any noticeable symptoms or reduction in fruit yield (Figure 1-2).

Another factor influencing symptoms in V. vinifera varieties is what species of virus are

present. Grapevine leafroll disease is one of the most complex viral infections known to infect plants (Jarugula et al. 2010) with the complex consisting of 10 different species of virus, named grapevine leafroll-associated virus-1 (GLRaV-1) through GLRaV-10 in the order of their discovery. We now know, however, that GLRaV-8 was erroneously described and should not be considered a valid species. All of these virus species are in the family Closteroviridae, characterized as being positive sense, single stranded RNA

(ssRNA) virions that are rod-shaped, 1,250-2,200 nm in length, with genome sizes of

15.5-19.3 kb (Martelli et al. 2002). The majority of the viruses in this complex (GLRaV-

1, -3, -4, -6, and -9) belong to the genus Ampelovirus while GLRaV-2 belongs to the genus Closterovirus (Jarugula et al.

2010) and GLRaV-7 is currently not assigned to any genus within the Closteroviridae.

Based on the sequence of data of four virus genomes within this

3 A B

C D

Figure 1-2. Grapevine leafroll disease in red-fruited varieties (A), white-fruited varieties

(B), Concord juice grapes (C), and its distribution within vineyards of V. vinifera (D)

(Source: A, B, and D, Naidu et al. 2008)

family (Agranovsky et al. 1994; Karasev et al. 1995; Klaassen et al. 1995; Jelkmann et al.

1997) the closterovirus genome has been divided into four modules (Dolja et al. 1994):

the accessory module, the core module, the molecular chaperone module, and the

structural module which can be divided into two conserved gene blocks, the

replicase block that is involved in virus replication and the quintuple gene block that is responsible for virion assembly, cell-to-cell and systemic movement of the virus, and

RNA silencing suppression (Dolja et al. 2006) (Figure 1-3). While genome

Replicase gene block Quintuple gene block

Figure 1-3. Genome organization of GLRaV-3

organization is a primary character for virus taxonomy, the type of vector that transmits a

given virus is also considered in taxonomic analyses (Karasev 2000).

Vector and Pathogen Interactions

In many instances, a genus of plant virus is transmitted by one family of insect and shares the same mode of transmission, suggesting a long evolutionary relationship between vector and pathogen (Nault 1997). This has also been confirmed by

Karasev (2000) and Tsai et al. (2010). However, within the

Closteroviridae, it does not appear that virus transmission is restricted to one family of insect vector, but possibly to superfamilies, exemplified by transmission of Ampelovirus species by insect species in both the Coccidae and Pseudococcidae. The specificity of mode of transmission to vector groups pertains to the location in the insect where the viral particles bind or are located and the amount of time it takes to acquire the virus

(acquisition access period) and transmit the virus (inoculation access period). There are three different modes of transmission with regard to insect vectors that have piercing, sucking mouthparts: non-persistent, semi-persistent, and persistent (Table 1-1). Viruses in the Closteroviridae are thought to

5 Table 1-1. Modes of Transmission of insect vectors with piercing-sucking mouthparts that transmit viruses

Mode of Transmission Acquisition Retention Location in Vector

Non-persistent Seconds to minutes Minutes to hours Stylet of mouthparts

Semi-persistent Minutes to hours Hours to days Foregut

Persistent

Circulative Hours to days Days to weeks Hemocele: virus will pass through gut tissue into the body cavity

and migrate to salivary glands

Propagative Hours to days Weeks to months Hemocele: virus will pass through gut tissue, replicate in the

body cavity, and migrate to salivary glands and/or ovaries

6 be transmitted in a semi-persistent manner. There is, however, some evidence of persistent transmission, as suggested by the presence of virions in the salivary glands (Cid et al. 2007). While vectors of GLRaV-2 have not yet been discovered, other viruses in this genus are known to be vectored by aphids (Karasev 2000).

Species of virus in the genus Ampelovirus, however, are vectored by mealybugs

(Pseudococcidae) and soft scale insects (Coccidae).

Grapevine Leafroll Disease in Washington State

Grapevine leafroll-associated viruses were first detected in Washington by Martin et al. (2005) as the result of a survey that was conducted to find viruses that had been present in California vineyards. Symptoms of GLRD had been observed in Oregon and

Washington vineyards in previous years and verification of the disease was necessary to make further management decisions. Currently there are 12 viruses infecting grapevines in Washington State (Table 1-2), although, the most serious viral infections are a result of the grapevine leafroll-associated viruses. The majority of research focusing on viral infections in Washington has been conducted on V. vinifera cvs.; research has largely ignored Vitis x labruscana L. The first record of virus in Vitis x labruscana L. was the detection of GLRaV-3 in 2006 (Soule et al. 2006). To date, no survey has been conducted to search for viruses in Vitis x labruscana L. vineyards. As of 2010, it was estimated that about 9% of Washington States wine grape acreage is affected by GLRD

(http://wine.wsu.edu/research-

7 Table 1-2. All species of virus detected in Washington State vineyards to date

Virus Species Year Detected Reference

GLRaV-1 2005 Martin et al. 2005

GLRaV-2 2005 Martin et al. 2005

GLRaV-3 2005 Martin et al. 2005

GLRaV-4 2006 Naidu et al. 2006

GLRaV-5 2006 Naidu et al. 2006

GLRaV-9 2008 Jarugula et al. 2008

GRSPaV 2005 Martin et al. 2005

GV-A 2010 Naidu, unpublished data

GV-B 2010 Naidu, unpublished data

GFLV 2008 Mekuria et al. 2008

GFkV 2010 Naidu and Mekuria 2010

GSyV-1 2010 Mekuria and Naidu 2010

extension/files/2010/05/crop-profile.pdf). However, with the wide distribution of natural vectors, it is likely that more acreage is infected. While GLRaVs are graft-transmissible,

GLRD in Washington State vineyards exists in clumps at the edge of vineyards (Naidu, unpublished data). This edge effect is an indication that GLRaVs are transmitted naturally by vectors in the field rather than introduction of infected plant material. The introduction GLRD to Washington was certainly a result of planting of infected plant material, however, since the introduction of the National Clean Plant Network (NCPN) in

2008 through The Farm Bill – H.R. 6124 Food, Conservation, and Energy Act of 2008, it

8 is unlikely that infected material has been introduced since the implementation of this act

and that natural spread of the virus in Washington is due to vectors transmitting virus

from infected plants that were brought prior to this act. Under this program, partnerships

between clean plant centers establish a set of guidelines and procedures that ensures that

all new propagative material is tested and verified as virus free.

Insect Vectors of Grapevine Leafroll Disease

The taxonomic group responsible for vectoring most GLRaVs is the Coccoidea.

The Coccoidea is a monophyletic group of phloem feeding insects that is commonly seen

as a pest in agricultural crops and ornamental plants, first appearing between 150 and 175

million years ago (Grimaldi and Engel 2005). Around the early Cretaceous, with the

diversification of angiosperms, the Coccoidea diversified and has since evolved into

about 7,700 extant species (Grimaldi and Engel 2005). This group of insects is the sister

group of the Aphidoidea, an evolutionary relationship that is reflected in transmission of

closteroviruses, where

species of the genus Closterovirus are transmitted by aphids and species of the genus

Ampelovirus transmitted by coccoids. This pattern of specific taxa vectoring viruses of a

given genus appears to be consistent across not only the scales and aphids, but also

whiteflies and is consistent enough to base virus taxonomy on which taxa they are

vectored by (Tsai et al. 2010). Even though GLRaV-2 has no known vectors, it is in the

genus Closterovirus and this close relationship is an indication of a long

evolutionary relationship between the virus and vectors. Various species of scales

and mealybugs can vector various species of virus and some viral species can be

9 transmitted by multiple species of vector (Table 1-3).

Of all the potential vectors of the various species of GLRaVs, the two that are

established and wide spread in Washington vineyards are the grape mealybug,

Pseudococcus maritimus (Ehrhorn), and the european fruit lecanium scale,

Parthenolecanium corni (Bouché). Pseudococcus maritimus has been known in

Washington vineyards since 1950 (Frick 1952) and is the predominant species of mealybug in vineyards while Pa. corni has not been officially recorded from the state of

Washington according to ScaleNet (2012). Despite the absence of an official record of this insect in the state, its widespread distribution across the state suggests that is has been present for a significant amount of time. Both species of insect have been demonstrated to be vectors of GLRaVs (Golino et al. 2002; Hommay et al. 2008)

between plants of V. vinifera. Due to, apparently, latent infections of GLRaVs in Vitis x labruscana L., no work has been done to see if Pa. corni and Ps. maritimus can successfully transmit GLRaVs intraspecifically with regard to Vitis x labruscana L. or interspecifically with regard to Vitis x labruscana L. and V. vinifera. This lack of information in the epidemiology of GLRD in Washington exposes a variable that could potentially be contributing significantly to the spread of GLRD. In Washington State, many vineyards of Vitis x labruscana L. and V. vinifera are grown in close proximity and in some case, blocks of Vitis x labruscana L. are immediately adjacent to blocks of V.

vinifera (Figure 1-4). Other factors that contribute to the epidemiology of GLRD is the life cycle of the insect vectors, effectiveness of monitoring, and management of vectors.

Both vectors have

10 Table 1-3. Species of GLRaVs and their associated vectors

Virus Vector Species Reference GLRaV-1 Heliococcus bohemicus1 Sforza et al. 2003 Parthenolecanium corni2 Sforza et al. 2003 Phenacoccus aceris1 Sforza et al. 2003 Pulvinaria innumerabilis2 Fortusini et al. 1997 Pulvinaria vitis2 Fortusini et al. 1997 GLRaV-2 Unknown GLRaV-3 Ceroplastes rusci2 Mahfoudhi et al. 2009 Heliococcus bohemicus1 Zorloni et al. 2006 Phenacoccus aceris1 Sforza et al. 2003 Planococcus citri1 Cabaleiro and Segura 1997 Planococcus ficus1 Engelbrecht and Kasdorf Pseudococcus 1990 calceolariae1 Petersen and Charles 1997 Pseudococcus comstocki1 Nakano et al. 2003 Pseudococcus longispinus1 Petersen and Charles 1997 Pseudococcus maritimus1 Golino et al. 2002 Pseudococcus viburni1 Golino et al. 2002 Pulvinaria innumerabilis2 Fortusini et al. 1997 Pulvinaria vitis2 Belli et al. 1994 GLRaV-4 Unknown GLRaV-5 Ceroplastes rusci2 Mahfoudhi et al. 2009 Pseudococcus longispinus1 Golino et al. 2002 GLRaV-6 Unknown GLRaV-7 Unknown GLRaV-9 Pseudococcus longispinus1 Sim et al. 2003 GLRaV-10 Unknown 1Pseudococcidae

2Coccidae

highly cryptic life styles, where adult females live in concealed locations such as under

bark and in crevices. Parthenolecanium corni is a parthenogenic species with three larval instars that under Washington growing conditions has one generation per year. It is small, ovoid, and brown (Figure 1-5). Pseudococcus maritimus reproduces sexually, and

under Washington conditions has two generations per year. The life cycle of Ps.

maritimus includes an egg stage, three (female) or four

11 V. v in i fe r a V. x labruscana L.

Figure 1-4. Vitis vinifera vineyard grown adjacent to a Vitis labruscana vineyard

(male) larval instars, and (in males) a pupal stage directly before the adult stage. The adult female of Ps. maritimus is neotenic and produces a waxy secretion that covers the entire body. The adult male is ephemeral, non-feeding, winged, and lacks the waxy secretion (Figure 1-5). For both vector species, it is the first instar that is of concern both for spread of GLRD and as a target for management practices. Adult stages of Pa. corni do not move once they have begun feeding and adult females of Ps. maritimus will migrate to different regions of the vine throughout the year. The first instar crawlers are the lifestage responsible for dispersal, making them the primary lifestage responsible for virus transmission. Since these lifestages are mobile and often move into more exposed regions than the adults, they are the target of spray insecticides.

12 A BC

Figure 1-5. Adult female of Pa. corni (A), adult female of Ps. maritimus (B), and adult

male of Ps. maritimus (C).

Due to the ability of both vectors to transmit GLRaVs between plants of V.

vinifera, the presence of GLRaVs in V. x labruscana L. and V. vinifera vineyards in

Washington State, proximity of V. x labruscana L. and V. vinifera, an edge-effect with regard to virus distribution in vineyards, the presence of both Pa. corni and Ps. maritimus in vineyards of V. x labruscana L. and V. vinifera, and lack of vector and virus management practices in V. x labruscana L. vineyards are indicators that V. x labruscana

L. vineyards could be serving as natural reservoir of GLRaVs ultimately contributing significantly to the spread of GLRD in Washington State.

13 References

Agranovksy, A. A., E. V. Koonin, V. P. Boyko, E. Maiss, R. Frötschl, N. A. Lunina,

and J. G. Atabekov. 1994. Beet yellows closterovirus: complete genome

structure and identification of a leader papain-like thiol protease. Virology.

198(1): 311-324.

Ball, T., R. J. Folwell, J. Watson, and M. Keller. 2002. Establishment and Annual

Production Costs for Washington Concord Grapes. Washington State University

Extension Bulletin EB

Belli, G., A. Fortusini, P. Casati, L. Belli, P. A. Bianco, and S. Prati. 1994.

Transmission of a grapevine leafroll associated closterovirus by the scale insect

Pulvinaria vitis L. Rivista di Patologia Vegetale. 4(3): 105-108.

Cabaleiro, C., and A. Segura. 1997. Field Transmission of Grapevine Leafroll

Associated Virus 3 (GLRaV-3). Plant Disease. 81: 283-287.

Cid, M., S. Pereira, C. Cabaleiro, F. Faoro, and A. Segura. 2007. Presence of

Grapevine leafroll-associated virus 3 in primary salivary glands of the mealybug

vector Planococcus citri suggests a circulative transmission mechanism.

European Journal of Plant Pathology. 118: 23-30.

14 Dolja, V. V., A. V. Karasev, and E. V. Koonin. 1994. Molecular biology and evolution

of closteroviruses: sophisticated build-up of large RNA genomes. Annual Review

of Phytopathology. 32: 261-285.

Dolja, V. V., J. F. Kreuze, and J. P. T. Valkonen. 2006. Comparative and functional

genomics of closteroviruses. Virus Research. 117: 38-51.

Engelbrecht, D. J., and G. G. F. Kasdorf. 1990. Transmission of grapevine leafroll

disease and associated closteroviruses by the vine mealybug, Planococcus ficus.

Phytophylactica. 22: 341-346.

Fortusini, A., G. Scattini, S. Prati, S. Cinquanta, and G. Belli. 1997. Transmission of

grapevine leafroll virus 1 (GLRaV-1) and grapevine virus A (GVA) by scale

insects. Proceedings of 12th Meeting of ICVG, Lisbon. 121-122.

Frick, K. E. 1952. The value of some organic phosphate insecticides in control of grape

mealybug. Journal of Economic Entomology. 45: 340-341.

Fuchs, M. F. 2007. Grape leafroll disease. Cornell University Integrated Pest

Management Extension Bulletin.

www.nysipm.cornell.edu/factsheets/grapes/diseases/grape_leafroll.pdf 2007

Jarugula, S., O. J. Alabi, R. R. Martin, and R. A. Naidu. 2010. Genetic Variability of

15 Natural Populations of Grapevine leafroll-associated virus 2 in Pacific Northwest

Vineyards. Phytopathology. 100: 698-707.

Harmon, F. N., and E. Snyder. 1946. Investigations on the occurrence, transmission,

spread, and effect of “white” fruit color in the Emperor grape. Proceedings of the

American Society of Horticultural Science. 47: 190-194.

Hommay, G., V. Komar, O. Lemaire, and E. Herrbach. 2008. Grapevine virus A

transmission by larvae of Parthenolecanium corni. European Journal of Plant

Pathology. 121: 185-188.

Golino, D. A., S. T. Sim, R. Gill, and A. Rowhani. 2002. California mealybugs can

spread grapevine leafroll disease. California Agriculture. 56(6): 196-201.

Jarugula, S., M. J. Soule, A Rowhani, and R. A. Naidu. 2008. First report of

Grapevine leafroll-associated virus 9 in Washington State vineyards. Plant

Disease. 92: 485.

Jelkmann, W., B. Fechtner, and A. A. Agranovsky. 1997. Complete genome structure

and phylogenetic analysis of little cherry virus, a mealybug-transmissible

closterovirus. Journal of General Virology. 78(8): 2067-2071.

Karasev, A. V., V. P. Boyko, S. Gowda, O. V. Nikolaeva, M. E. Hilf, E. V. Koonin,

C. L. Niblett, K. Cline, D. J. Gumpf, R. F. Lee, S. M. Garnsey, D. J.

Lewandowski, and W. O. Dawson. 1995. Complete Sequence of the Citrus

Tisteza Virus RNA Genome. Virology. 208(2): 511-520.

16 Karasev, A. V. 2000. Genetic Diversity and Evolution of Closteroviruses. Annual

Review of Phytopathology. 38: 293-324.

Klaassen, V. A., M. L. Boeshore, E. V. Koonin, T. Tian, and B. W. Falk. 1995.

Genome Structure and Phylogenetic Analysis of Lettuc Infectious Yellow Virus, a

Whitefly-Transmitted, Bipartite Closterovirus. Virology. 208(1): 99-110.

Martelli, G. P., A. A. Agranovsky, M. Bar-Joseph, D. Boscia, T. Candresse, R. H. A.

Coutts, V. V. Dolja, B. W. Falk, D. Gonsalves, W. Jelkmann, A. V. Karasev,

A. Minafra, S. Namba, H. J. Vetten, G. C. Wisler, and N. Yoshikawa. 2002.

The family Closteroviridae revised. Archives of Virology. 147: 2039-2044.

Martin, R. R., K. C. Eastwell, A. Wagner, S. Lamprecht, and I. E. Tzanetakis. 2005.

Survey for Viruses of Grapevine in Oregon and Washington. Plant Disease. 89:

763-766.

Mekuria, T., R. R. Martin, and R. A. Naidu. 2008. First Report of the Occurrence of

Grapevine fanleaf virus in Washington State Vineyards. Plant Disease. 92(8):

1250.

Mekuria, T., and R. A. Naidu. 2010. First Report of Grapevine Virus Sequences Highly

Similar to Grapevine Syrah virus-1 from Washington Vineyards. Plant Disease.

94(6): 787.

17 Myles, S., A. R. Boyko, C. L. Owens, P. J. Brown, F. Grassi, M. K. Aradhya, B.

Prins, A. Reynolds, J. Chia, D. Ware, C. D. Bustamante, and E. S. Buckler.

2011. Genetic structure and domestication history of the grape. Proceedings of

the National Academy of Sciences. 108(9): 3530-3535.

Naidu, R. A., M. J. Soule, and S. Jarugula. 2006. Single and mixed infections of

Grapevine leafroll-associated viruses in Washington State vineyards.

Phytopathology. 96: S83.

Naidu, R. A., S. O’Neal, and D. B. Walsh. 2008. Grapevine leafroll disease.

Washington State University Bulletin EB2027E.

Naidu, R. A., and T. Mekuria. 2010. First Report of Grapevine fleck virus from

Washington Vineyards. Plant Disease. 94(6): 784.

Nakano, M., R. Nakaune, and S. Komazaki. 2003. Mealybug transmission of

grapevine viruses in Japan. Proceedings of the International Council to Study

Viruses and Virus-like Diseases of Grapevine, Locorotondo, Italy. 218.

Nault, L. R. 1997. Arthropod transmission of plant viruses: A new synthesis. Annals of

the Entomological Society of America. 90: 521-541.

18 Petersen, C. L., and J. G. Charles. 1997. Transmission of grapevine leafroll-associated

closteroviruses by Pseudococcus longispinus and P. calceolariae. Plant

Pathology. 46: 509-515.

Sforza, R., E. Boudon-Padieu, and C. Greif. 2003. New mealybug species vectoring

Grapevine leafroll-associated viruses-1 and -3 (GLRaV-1 and -3). European

Journal of Plant Pathology. 109: 975-981.

Sim, S. T., A. Rowhani, and D. A. Golino. 2003. Experimental transmission of

grapevine leafroll-associated virus 5 and 9 by longtailed mealybugs. Proceedings

of the International Council to Study Viruses and Virus-like Diseases of

Grapevine, Locorotondo, Italy. 211-212.

Soule, M. J., K. C. Eastwell, and R. A. Naidu. 2006. First Report of Grapevine leafroll

associated virus-3 in American Vitis spp. Grapevines in Washington State. Plant

Disease. 90(11): 1461.

Stonebridge Research. 2012. Economic Impact of Washington State Wine and Grapes.

Stonebridge Research Report for Washington State Wine Commission. p.1-53.

Walker, J. T. S., J. G. Charles, K. J. Froud, and P. Connolly. 2004. Leafroll virus in

vineyards: modeling the spread and economic impact. HortResearch Client

Report No. 12795.

19 Zorloni, A., S. Prati, P. A. Bianco, and G. Belli. 2006. Transmission of grapevine virus

A and grapevine leafroll-associated virus 3 by Heliococcus bohemicus. Journal of

Plant Pathology. 88(3): 325-328.

20 CHAPTER 2

PHEROMONE-BASED MONITORING OF PSEUDOCOCCUS MARITIMUS

(HEMIPTERA: PSEUDOCOCCIDAE) POPULATIONS IN CONCORD GRAPE

VINEYARDS

Introduction

The grape mealybug, Pseudococcus maritimus (Ehrhorn), was first reported as a pest in

Washington State in 1950 (Frick 1952). In Wa shington, Ps. maritimus populations have

been reported to infest a variety of fruits including apricot (Madsen and McNelly 1960), grape (Frick 1952, Cone 1971), and pear (Doutt and Hagen 1950). Historically, economic losses in vineyards were caused by the mealybugs’ honeydew, which provided a substrate for the growth of sooty mold and other food contaminants, or direct infestation of the grape clusters by mealybugs (Daane et al. 2012). However, Ps. maritimus is capable of transmitting grapevine leafroll-associated viruses (GLRaVs; Golino et al. 2002, Tsai et al.

2011), which are the causal agents of grapevine leafroll disease (GLRD; Rayapati et al.

2008, Golino et al. 2002), the most important viral disease of grapevines worldwide

(Martelli et al. 2002), thus elevating the economic consequences of Ps. maritimus

infestations.

Pseudococcus maritimus females have three larval instars and an adult stage,

whereas males have three larval instars, a pupal stage (fourth instar), and a winged adult

stage (Grimes and Cone 1985b). Stages present on the vine and locations and densities of

mealybugs vary seasonally (Daane et al. 2012). Pseudococcus maritimus populations

typically complete two generations per year, overwintering under the bark of the trunk,

21 cordon, or spurs as eggs in the cotton-like ovisac, or as first instar crawlers (Geiger and

Daane 2001, Grasswitz and James 2008). With warming spring temperatures, part of the population moves upward to feed on the canes and leaves for several weeks and then moves back to protected locations under the bark to complete development. Gravid females also deposit ovisacs under loose bark. Eggs from the second generation hatch in midsummer, and the immature Ps. maritimus repeat the upward movement on the vine, with the mealybugs commonly infesting grape clusters touching the vine trunk or spurs

(Geiger et al. 2001). As temperatures cool in the fall, much of the mealybug population moves back to protected locations and mated females from this summer generation oviposit under the bark, establishing the next overwintering generation.

Historically, management of Ps. maritimus in Washington’s Concord cv. grapes relied on applications of broad-spectrum organophosphate insecticides (Grimes and Cone

1985a), many of which have lost their efficacy due to the development of resistance

(Flaherty et al. 1982), or are no longer registered for use in vineyards. Currently, the predominant Ps. maritimus treatment in Washington wine grapes involves chemigation, whereby imidacloprid is applied through the drip system and the vines take up the material systemically (Daane et al. 2006). However, most of Washington’s Concord cv. vineyards are irrigated by overhead sprinklers, eliminating the option of chemigation, and thereby leaving foliar sprays as the primary control tool. With some foliar materials, application timing is important to target immature mealybugs (crawlers through second instars) that have less of the protective waxy coating than adults, and which are also mobile, allowing them to get more exposure to the applied insecticide(s). The efficacy of foliar insecticides can be further limited because the dense canopy structure in many

22 Concord cv. vineyards reduces insecticide coverage and because of the seasonal

movement of mealybugs from exposed to more protected locations on the vine described

above.

For the above reasons, monitoring for Ps. maritimus constitutes a crucial

component of vineyard integrated pest management (IPM) programs, to detect small

populations that could result in the transmission of grapevine leafroll-associated viruses

(GLRaVs), and to time foliar insecticide applications to the presence of immature stages.

However, developing comprehensive monitoring guidelines for Ps. maritimus has been

problematic because of the mealybugs’ cryptic nature, their clumped distribution within

vineyards, and their distribution on different parts of the vine during different times of the year (Geiger and Daane 2001). Sampling techniques for Ps. maritimus have included labor-intensive visual counts based on a unit area or time, which targeted both immature mealybugs and adult females, and sampling a number of randomly selected vines within a block (Grimes and Cone 1985b, Geiger and Daane 2001). Traditionally, adult males of

Ps. maritimus have not been sampled due in part to their small size and ephemeral presence. However, the recent identification of the Ps. maritimus sex pheromone

(Figadère et al. 2007, Zou et al. 2010) provided the means to develop a sensitive and selective method to sample male mealybugs with pheromone-baited traps.

The primary objectives of this study were to determine the seasonal flight activity of adult males of Ps. maritimus in Washington Concord cv. vineyards, and to determine optimal pheromone trap densities in vineyards to use in a Ps. maritimus monitoring program. A secondary objective was to use pheromone traps to monitor for the presence of three mealybug species known to be present in western U.S. vineyards but not yet

23 recorded from vineyards in Washington. These are Planococcus ficus (Signoret),

Pseudococcus longispinus (Targioni Tozzetti), and Pseudococcus viburni (Signoret)

(Daane et al. 2008).

Materials and Methods

Field Test Sites. Experiments were conducted during the 2009 to 2011 growing

seasons in commercial Concord cv. vineyards located near the Washington State

University Irrigated Agriculture Research and Extension Center (IAREC) in Prosser,

WA. These vineyards were managed with standard commercial practices, with the

exception that no insecticides were applied against Ps. maritimus. They are referred to by

vineyard site or owner as Alguist (Alq), Boast (Boa), Cabbage (Cab), Daves (Dav),

Hogue (Hog), Little (Lit), New Vineyard (NV9), Sunridge (Sun), and Vanhinkin (Van).

Pheromone-Baited Traps. In all trials, sex pheromones for Ps. maritimus (Zou et al. 2010), Ps. viburni (Millar and Midland 2007), Ps. longispinus (Millar et al. 2009), and

Pl. ficus (Hinkens et al. 2001) were synthesized as previously described and loaded on 11 mm grey silicon rubber septa (West Pharmaceutical Services, Lititz, PA) as hexane solutions (25 µg of racemic pheromone in 25 µl of hexane), separately for each species.

Pheromone lures were deployed in delta sticky traps (Suterra Inc., Bend, OR) and placed in the upper vine canopy as described by Walton et al. (2004).

Trap Density. In 2009, 10 to 13 traps baited with Ps. maritimus pheromone were placed in each of 12 vineyards to help locate sites with measurable mealybug densities.

Each site was large enough to be divided into multiple 12.14 ha plots. Traps were deployed from 23 June to 1 October 2009; traps were collected and adult male mealybugs were counted weekly, with the pheromone lures replaced every 6 to 8 wk.

24 From the vineyards sampled in 2009, seven vineyards with measurable Ps. maritimus populations were selected for study in 2010. Each vineyard was divided into three 12.14 ha plots. Traps were deployed at densities of one, four, or eight Ps. maritimus pherom one-baited traps per 12.14 ha. Traps in the four- and eight-trap density treatments were placed 10 vines apart in a single row. Traps were deployed from 20 May to 27

October 2010. Each vineyard had all three trapping densities set in a randomized block design, with each vineyard serving as a replicate.

Additionally, in 2010, traps baited with the pheromones of Ps. viburni, Ps. longispinus, Pl. ficus, or blank control lures were deployed in each of the seven sampled vineyards, with one trap per species (or control) per vineyard. Any mealybugs captured in these traps were collected, stored at -80°C and later identified using multiplex PCR methods, as described by Daane et al. (2011).

In 2011, nine vineyards were selected for study, each with one 12.14 ha plot. Each plot had only one trap density (one, four, or eight), with treatments established in a completely randomized design amongst the nine vineyards, resulting in three replicate vineyards for each trap density. Traps were deployed from 12 April to 24 October 2011.

Data Collection and Analysis. Trapped male mealybugs were counted using a

Leica MZ7.5 dissecting microscope (Leica, Wetzlar, Germany). Data were analyzed using Systat 12 (Systat Software, Inc. 2007) and STATGRAPHICS Plus Version 5.0

(Statistical Graphics Corp. 2000). Data are presented as means (± SE). Analysis of variance (ANOVA) was used to determine treatment effects on each sample date and differences in average trap captures among vineyard sites, with vineyards modeled as a blocking factor, and treatment means separated using Tukey’s pairwise comparisons (P <

25 0.05). Trap data were also analyzed to determine treatment effects on the mean number of

adult male Ps. maritimus captured per trap among trap densities over time (season-long

capture patterns) using Repeated Measures ANOVA.

Sample Size. A number of formulae have been used to express a population’s

spatial distribution, typically based on the variance to mean relationship from collected

samples of the population (Karandinos 1976, Young and Young 1998). Taylor (1961)

related the variance and mean as: s2 = axb, where s2 is the sample variance and x is the sample mean, which is solved with the linear regression of the natural logarithms: ln(s2) = ln(a) + blog(x). The slope b, or Taylor’s coefficient, also describes the population’s distribution pattern: If b > 1, the distribution is aggregated or clumped (the data best fit a negative binomial distribution); if b < 1 the distribution is uniform (the data are best fit to a binomial distribution), and if b = 1 the population has a random distribution (the data fit a Poisson distribution) (Young and Young 1998). Iwao’s patchiness regression (Iwao

1968) is based on Lloyd’s mean crowding index (m = x + s2 / x – 1) and the regression

model is m = α + βx, where the slope (β) has the same meaning as Taylor’s coefficient

(Ifoulis and Savopoulou-Soultani 2006). Optimal sample size (Nopt), referring to trap

number, was then determined using Taylor’s coefficient:

2 b-2 Nopt = (Zα/2 / C) (ax ) (1) and Iwao’s patchiness regression:

2 Nopt = (Zα/2 / C) ((α +1)/x + (β – 1)) (2)

where Zα/2 is the upper α/2 of the standard normal distribution, α is the set confidence

level, and C is the precision error. Here we used a 95% confidence interval so that Zα/2 is

1.96, after Mallampalli and Isaacs (2002), and a precision level of 30%. Optimal sample

26 size was also determined using the K parameter of the negative binomial model (Young

and Young 1998):

2 Nopt = (Zα/2/C) (1/x + 1/ Kc) (3)

where Kc is the common parameter of the negative binomial model. Individual K values

were determined from variance and mean data for each sample date and vineyard, as K =

2 2 x / (s + x) and Kc is the average of the K values. Bacca et al. (2008) used a derivation of eq. (3) to determine sample size for a fixed mean population and varying precision levels

as:

2 Nopt = (1/C ) (1/x + 1/ Kc) (4)

For all sample size analyses, mean and variance data were derived from the 2010

collections within each vineyard, which provided 13 trap counts (one, four and eight traps

per 12.14 ha treatments) in each of seven vineyards for each of 23 sample dates. The

2010 data were also analyzed within each treatment, which provided data for treatments of one (seven traps), four (28 traps), and eight (56 traps) traps per 12.14 ha plot per sample date. For this analysis, the mean of means was used to reduce the impact of high variability in mealybug counts among vineyards, and prevent pseudo-replication

(resulting in seven replicates for each sample date).

Results

Seasonal Phenology. Over 23 sample dates in 2010, a total of 168,092 adult male

Ps. maritimus were recorded from 2,090 pheromone trap-weeks, with an average of 80.4

± 3.6 Ps. maritimus per trap per week. Over 28 sample dates in 2011, total and average

counts were lower, with 32,435 adult male Ps. maritimus recorded from 1,092 trap-

27 weeks, with an average of 29.7 ± 2.2 Ps. maritimus per trap per week. There was a significant difference among vineyards in the average trap catch, across all sampling dates, in 2010 and 2011 (Fig. 1). Seasonal phenology was similar in both years of the study (Fig. 2). In 2010, Ps. maritimus was found in low numbers during the first week that traps were deployed (20 to 27 May) in all vineyards sampled. The peak of the first flight occurred during the week of 3 to 10 June 2010, and the second flight peaked from 5 to 25 August followed by a small decline and a smaller peak in the second week of

September (Fig. 2). In 2011, traps were deployed earlier (12 April) in order to catch the beginning of the first flight, which was detected from 2 to 10 May. The peak of the first flight occurred between 6 and 27 June 2011; the peak of the second flight was between

22 and 29 August, with a smaller peak following in the second week of September (Fig.

2).

Two distinct flight periods were recorded in both years, with a relatively similar seasonal phenology, although peak flights occurred earlier in 2010 than in 2011.

Individual data loggers were not placed in each block, but degree day accumulation, based on Washington State University’s AgWeatherNet (http://www.weather.wsu.edu) for grape development, showed the accumulated day-degrees (base 10°C) at IAREC were low (2,325 DD) in the 2010 season as compared with the long term average of 2,526 DD.

Degree day accumulation was even lower in 2011 (2,312 DD).

Trap Density Comparisons. In 2010, there was no season-long difference in the pattern of adult male captures among the three trapping density treatments over the 23 wk sample period (Repeated Measures ANOVA: F = 0.72; df = 2, 18; P = 0.50; Fig. 3a).

When data were analyzed on a weekly basis, using the mean per vineyard for each

Figure 2-1. Mean numbers (±SE) of adult male Pseudococcus maritimus recorded per

trap per week in (A) 2010 (23 sample dates) and (B) 2011 (28 sample dates) show a

significant difference among 12.14 ha vineyard blocks sampled (ANOVA for 2010: F =

18.63, df = 6, 2885, P < 0.001; 2011: F = 16.32, df = 8, 1083, P < 0.001). The vineyards

corresponding to the abbreviations shown on the x-axis are listed in the methods section.

trapping density (seven replicates per treatment), there were no significant treatment effects among the one, four, or eight traps per 12.14 ha treatments on any of the 23 sampling dates.

In 2011, no season-long difference in the pattern of adult male captures was found among trap density treatments (Repeated Measures ANOVA: F= 0.91; df =2, 6; P=0.45;

Fig. 3b). When data were analyzed on a weekly basis, using the mean per vineyard for each trapping density (three replicates per treatment), there were significant treatment effects in wk 6, when more adult male Ps. maritimus were recorded from plots with four traps per 12.14 ha over the other treatments (ANOVA: F= 28.00; df = 2, 6; P = 0.001), and week 12 when more males were recorded from plots with four traps per 12.14 ha

29 600 eek

500 2010 2011 400 ) per trap per w 300

200

100

0 Male mealybugsMale (±SE Apr May Jun Jul Aug Sept Oct Nov Dec

Figure 2-2. Mean numbers (±SE) of adult male Pseudococcus maritimus recorded per trap per week in 2010 and 2011 show two distinct flight periods in each year.

than the plots with eight traps per 12.14 ha (ANOVA: F= 5.77; df = 2, 6; P = 0.040).

During the 2011 season, when treatments were in different vineyards rather than in each vineyard, there was greater variation in male mealybug counts among treatments (Fig.

3b). When all trap data were used (three, 12, and 24 traps per sample date for trap densities of one, four, and eight traps per 12.14 ha, respectively), traps at a density of eight per 12.14 ha consistently had lower trap catches than traps at densities of one and four per 12.14 ha (Fisher LSD, P < 0.05).

Sample Size. The relationship between ln variance and ln mean was significant for adult male Ps. maritimus captures (Fig. 4). The slope (b), Taylor’s coefficient, was significantly >1 for male captures when data were analyzed by vineyard (Student’s t-test; b = 1.53 ± 0.537; t = 61.62, P < 0.001) or by treatment (Student’s t-test; b = 1.783 ±

Figure 2-3. Mean numbers (±SE) of adult male Pseudococcus maritimus recorded per trap per week for trapping densities of one, four, and eight traps per 12.14 ha in (A) 2010 and (B) 2011.

0.032; t = 55.57, P < 0.001). These data indicate that pheromone trap captures of adult male Ps. maritimus are well described by the negative binomial distribution, and that the population is in an aggregated pattern. The mean to variance data analyzed by vineyards provided a lower slope and was therefore used to determine optimal sample size for all equations used.

Relationships between the optimum number of samples (pheromone traps) per

12.14 ha sample block and mean number of captured adult male Ps. maritimus per trap per week (pest density), are shown for Taylor’s coefficient (eq. 1), Iwao’s regression parameters (eq. 2), and the Kc parameter of the negative binomial (eq. 3), using a 30% precision level (Fig. 5). The 30% precision lines began to level off at 20 adult males captured per trap using the K parameter of the negative binomial model, but not until 70 adult males were captured using Taylor’s and Iwao’s parameters, the latter two of which

Figure 2-4. Regression curve between the natural logarithm (ln) of the variance and the

ln of the mean (x) number of adult male Pseudococcus maritimus captured to obtain the

coefficients of Taylor’s power law for data collected in 2010 showing positive

relationship between ln variance and ln mean when data were analyzed by (A) vineyard

(y = 1.53x + 0.537; F = 3797, df = 1, 146; P < 0.001; r2 = 0.963)or (B) treatment (y =

1.783x + 0.444; F = 3087; df = 1, 65; P < 0.001; r2 0.979).

also required more traps for the same level of precision (Fig. 5). At 50 mealybugs per

trap, the required sample size began to level off, and was 11.6, 9.5 and 8.0 traps per 12.14

ha for Taylor’s coefficient, Iwao’s regression parameters, and the Kc parameter,

respectively. The complete data set included sample dates between flights when trap

captures averaged < 5 Ps. maritimus per trap per week (Fig. 2). Using the complete trap

data set, the Kc parameter was 5.97; sample dates with greater than 50 or 100 adult male

Ps. maritimus per trap per week better reflected trap efficiency during active flight

periods, resulting in Kc parameters of 8.92 and 10.50, respectively, and required sample sizes of 5.9 and 4.6 traps per 12.14 ha, respectively (Fig. 6a). All models suggested that

32 50

Taylor 40 Iwao K parameter

e e 30 mpl size

20 Optimal sa Optimal 10

0 01020 30 50 70 100

Male mealybugs per trap

Figure 2-5. Relationship between optimal sample size (pheromone traps per 12.14 ha

vineyard) and adult male Pseudococcus maritimus per trap per week for sample size

models based on Taylor’s Power Law (eq. 1) where a = 1.71 and b = 1.53, Iwao’s

regression parameters (eq. 2) where α = 2.16 and β = 1.19, and the K parameter of the negative binomial (eq. 3) where Kc = 5.97. Data were from 2010 trap captures, with mean

– variance relationships determined from each 13 trap data set collected in each of seven

sampled vineyards on 23 sample dates.

more than eight traps per 12.14 ha were needed for a 30% precision level, which would

be more than vineyard pest managers would consider economical. For this reason, we

further analyzed the data to determine if the optimal sampling number could be lowered,

still using the existing mean to variance relationship. The required number of traps for

varying precision levels (eq. 4, Bacca et al. 2008) stabilized at the 40% precision error at less than 1 trap per 12.14 ha, for Kc parameter values from trap data with greater than 50

Figure 2-6. (A) Relationship between optimal sample size (pheromone traps per 12.14 ha vineyard) and adult male Pseudococcus maritimus per trap per week using the K parameter of the negative binomial (eq. 3) where Kc = 5.97 (all 2010 trap data, as in Fig.

5), Kc = 8.92 (pheromone traps with >50 males per trap) and Kc = 10.50 (pheromone traps with >100 males per trap) and Kc = 10.50; and (B) number of pheromone traps

(samples) per 12.14 ha vineyard needed under different precision errors from 5 to 40% as determined using Kc (eq. 4) determined from all 2010 trap data, traps with > 50 males, and traps with >100 males per trap per sample date.

and 100 captured Ps. maritimus per trap (Fig. 6b). Increasing precision to the levels of

30% and 20% required 1.2 to 2.5 and 2.7 to 5.7 traps per 12.14 ha, respectively, for varying these same Kc values (Fig. 6b). Our initial analyses (Fig. 5) used season-long trap captures, including periods before, between, and after peak flight periods, when trap captures typically ranged from 0 to 3 males per trap per week (Fig. 2). At lower trap capture densities, more traps were needed to assess the population density, using any of

34 the optimal sampling models (Fig. 5). This was similarly shown for visual samples of Ps. maritimus on the vine (Geiger and Daane 2001). Most sampling programs show fewer samples are needed with increasing density of the sampled population (e.g., Mallampalli and Isaacs 2002, Ifoulis and Savopoulou-Soultani 2006) when the population fits the negative binomial model (Young and Young 1998). For this reason, optimal sample size was also determined using the Kc parameter for trap captures with >50 and >100 Ps. maritimus per trap. These were reasonable capture rates during the flight period; in traps with adult males (excluding trap captures with zero males), average trap capture was

107.4 ± 4.5 and 59.4 ±4.0 males per trap per week in 2010 and 2011. With this analysis, required sample size was reduced to five traps per 12.14 ha, with little difference between traps with >50 or >100 male captures (Fig. 6a). Bacca et al. (2008) also used the K parameter to determine optimal sample size of pheromone traps with varying precision levels and a fixed population size of the coffee leaf miner, Leucoptera coffeella (Guérin-

Méneville & Perrottet). Using this more simplified equation (Zα/2 is replaced by 1), Ps.

maritimus sample size was reduced to about one and two pheromone traps per 12.14 ha

for precision levels of 30 and 25%, respectively (Fig. 6b).

Mealybug species. In 2010, there were 12, 19, 12, and 24 male mealybugs

captured in traps baited with pheromones of Ps. viburni, Ps. longispinus, Pl. ficus, or a

solvent control (no pheromone), respectively. Multiplex PCR analysis determined that all

of these collected male mealybugs were Ps. maritimus.

Discussion

Pseudococcus maritimus was previously reported to have two generations per year in Washington vineyards, based on visual samples of life stages found on the vine

35 (Grimes and Cone 1985b, Geiger and Daane 2001). Here, we provide evidence of two distinct flight periods (Fig. 1). In 2011, the first male Ps. maritimus were caught on May

10; based on the assumption of a 10°C lower temperature developmental threshold for

Ps. maritimus (Geiger and Daane 2001), this corresponded to a 107 degree-day accumulation from 1 January 2011 to first capture. The beginning of the first flight was likely missed in 2010 because the first traps were deployed on 10 May (152 degree-day accumulation) and these traps captured several adult male Ps. maritimus, suggesting that the flight had begun before traps were deployed. The unusually low spring temperatures also may have been a factor in the reduced Ps. maritimus trap captures in 2011 compared with 2010, either as a result of reduced mealybug survival, or reduced flight activity by males. The two peaks in trap captures were in early June and mid-August, which match the seasonal periods when adult females are producing ovisacs (Grimes and Cone 1985b,

Geiger and Daane 2001). The first flight peak in 2011 occurred one week later than in

2010, and the second flight peak in 2011 occurred two weeks later than the second peak in 2010. This slight delay was most likely a result of below average temperatures in the spring of 2011 (WSU AgWeatherNet 2011) that also caused delays in the development of many fruit crops, including juice grapes, in central Washington.

Pheromones have long been used in insect pest management programs, primarily to monitor pest densities (Burkholder 1985, Suckling 2000, Way and van Emden 2000).

However, the use of semiochemicals for monitoring mealybugs is relatively new and/or rarely practiced. The pheromone of the citrus mealybug, Planococcus citri (Risso), has been known for more than 30 years (Bierl-Leonhardt et al. 1981), and whereas it has been used in some Mediterranean countries for monitoring (Franco et al. 2004), this practice

36 has not been widely adopted and we could find no guidelines for its use in control decisions. In vineyards, pheromone trapping for Pl. ficus in South Africa (Walton et al.

2006) and California (Millar et al. 2002) has been tested and pheromone lures are commercially available; however, Pl. ficus trapping in California has not yet seen widespread adoption because most managers apply insecticides as soon as any Pl. ficus are found in vineyards. In nursery systems with ornamental plants, Waterworth et al.

(2011a) recently tested pheromone-baited traps for monitoring Ps. viburni, Ps. longispinus, and Pl. citri and showed that trap counts of male mealybugs were correlated with mealybug species and densities on nearby plants, suggesting considerable potential for pheromone-based monitoring of mealybugs in nursery systems.

In this study, we provide baseline data to develop a pheromone-based monitoring program for Ps. maritimus in vineyards. We showed that trap catches (Fig. 2) match the known seasonal phenology for Ps. maritimus. The number of traps needed per vineyard block was determined using field trials with different trapping densities, and through analysis of 2010 trap capture data using Taylor’s Power Law, Iwao’s regression parameters, and the K parameter from the negative binomial model. In 2010, seasonal average trap captures were different among individual vineyards (Fig. 1a), but male Ps. maritimus captures per trap were similar across all treatments (Fig. 3a) and on individual sample dates. These data indicate that a single trap per 12.14 ha block (averaged across seven replicates) matches the results of four and eight traps per block densities. In 2011, trap density treatments were each placed in separate vineyards, and for this reason, seasonal average trap captures were different among individual vineyards (Fig. 1b), creating differences in average trap captures per sample date and density treatment (Fig.

37 3b). The more important measure in 2011 was seasonal phenology of male flight periods as determined from the trap catch data. Here, the traps performed well, with the first seasonal captures of adult males being similar across all treatments (four of nine blocks on 18 to 25 May), as were peak counts in the first flight (nine of nine blocks on either 14 to 21 May or 21 to 28 May) (Figs. 1 and 3).

Peak counts in the second flight were less synchronized as compared with those in the first flight, and there was a double peak in both study years. The double peak could result from a late developing second generation or a partial third generation of adult males emerging. Another possible explanation for the double peak is related to the changing age structure of the female population. The late August to early September decline in captures may result from the presence of sexually mature females producing pherom one that competed with the trap lures. Once these females were mated, there might be less competition from them, resulting in an increase in trap captures. However,

Waterworth et al. (2011b) showed that mated Ps. viburni and Ps. longispinus females, both close relatives of Ps. maritimus, remated multiple times, suggesting that mature females might still be producing pheromone. Thus, it seems less likely that the double peak in the pheromone trap catches was related to competition of lures with pheromone- producing females.

Three commonly used regression models were used to estimate optimal sample size: Taylor’s Power Law (TPL), Iwao’s patchiness regression (IPR), and the negative binomial model. For pest management, sampling precision levels between 20 to 35% are desirable, whereas for research, precision levels between 10 and 20% are often required.

While precision was set to the same value for each method, one cannot assume that this

38 desired level of precision would have been equally well achieved by the three models.

Also, the sample sizes tested in this study were observed trap densities that had been

employ ed in previous IPM programs. The three models used to estimate sample size were

used to determine if the current practices met a fixed level of precision in monitoring

populations of Ps. maritimus. While the nominal level of precision for research purpose

is attainable at higher sample sizes, especially when the adult male population is low, for

growers seeking a faster, more economical means for detecting Ps. maritimus to guide

their IPM programs, a lower sample size (1 trap per 12.14 ha.) is adequate. The purpose

of the pheromone-based monitoring programs for Ps. maritimus is first to determine

presence or absence of the pest and secondly to be able to determine what stage of

development the mealybugs are in and at what time of the growing season so as to guide

application of insecticide treatments so they are most effective (i.e. targeting 1st instar

crawlers). Whereas TPL and IPR may yield more precise estimates more frequently, the

less precise K model is adequate for IPM programs that seek to use insecticide

treatments, when efficacy is independent of population estimates, regardless of the level

of precision.

It is also important to emphasize the usefulness of pheromone traps in detecting

invasive mealybug species. In the Washington vineyards monitored, no Ps. viburni, Ps.

longispinus, or Pl. ficus were found in any of the vineyards. The numbers of males caught

in traps baited with pheromones of those three species were less than those caught in control traps, and conclusive identification by DNA analysis confirmed that all specimens caught were Ps. maritimus that had randomly flown into the traps.

39 Pheromone traps are commonly used to monitor first flight periods in order to set a biofix of population development or to determine peak population flight activity, which is often correlated to subsequent economic damage (Suckling 2000, Way and van Emden

2000). The use of pheromone traps at either one or two per 12.14 ha vineyard block is far less time consuming than visual searches of vines, which required 18 to 25 vines sampled for 5 min each to provide similar precision levels (Geiger and Daane 2001). Whereas trap capture s of adult male Ps. maritimus were clumped in distribution (b > 1), the winged adult males disperse over far greater distances than the sessile and highly clumped immature and female populations, thus requiring fewer samples. The determination of an optimal trap density that provides reliable information on pest populations while minimizing the costs of monitoring is an important aspect in the development of a vineyard IPM program. For example, to use insect growth regulating (IGR) insecticides effectively for Ps. maritimus control, growers need to know when crawlers, the targeted life stage, are most likely to be present. Extrapolating forward from peaks of flight activity by mate-seeking adult males, and using the number of degree-days required for eggs to be laid and to hatch, should allow accurate estimation of when to spray IGRs for maximum efficacy. For this type of monitoring, our results show that one trap per 12.14 ha of vineyard may be an effective tool to accurately detect adult male Ps. maritimus flight activity. Our results also demonstrate that pheromone-baited traps provide a sensitive method of detecting Ps. maritimus infestations. Future studies must be conducted to correlate trap densities to mealybug damage levels in vineyards (Walton et al. 2004, Ifoulis and Savopoulou-Soultani. 2006), and to determine optimal trap placement (Bacca et al. 2006). All of these factors will be critically important for better

40 management and control of Ps. maritimus and the leafroll viral pathogens that they transmit.

41 References

Bacca, T., E. R. Lima, M. C. Picanço, R. N. C. Guedes, and J. H. M. Viana. 2006.

Optimum spacing of pheromone traps for monitoring the coffee leaf miner

Leucoptera coffeella. Entomol. Exp. Appl. 119: 39-45.

Bacca, T., E. R. Lima, M. C. Picanço, R. N. C. Guedes, and J. H. M. Viana. 2008.

Sampling plan for the coffee leaf miner Leucoptera coffeella with sex pheromone

traps. J. Appl. Entomol. 132: 430–438.

Bierl-Leonhardt, B. A., D. S. Moreno, M. Schwarz, J. Fargerlund, and J. R.

Plimmer. 1981. Isolation, identification and synthesis of the sex pheromone of the

citrus mealybug, Planococcus citri (Risso). Tetrahedron Lett. 22: 389-392.

Burkholder, W. E. 1985. Pheromones for monitoring and control of stored-product

insects. Ann. Rev. Entomol. 30: 257-272.

Cone, W. C. 1971. Grape mealybug control in Concord grape field trials in central

Washington. J. Econ. Entomol. 64: 1552-1553.

Daane, K. M., W. J. Bentley, V. M. Walton, R. Malakar-Kuenen, J. G. Millar, C. A.

Ingels, E. A. Weber, and C. Gispert. 2006. New controls investigated for vine

mealybug. Calif. Agric. 60: 31-38.

42 Daane, K. M., M. L. Cooper, S. V. Triapitsyn, V. M. Walton, G. Y. Yokota, D. R.

Haviland, W. J. Bentley, K. E. Godfrey, and L. R. Wunderlich. 2008. California

vineyard managers and researchers seek sustainable solutions for mealybugs, a

changing pest complex. Calif. Agric. 62: 167-176.

Daane, K. M., M. C. Middleton, R. Sfroza, M. L. Cooper, V. M. Walton, D. B.

Walsh, T. Zaveizo, and R. P. P. Almeida. 2011. Development of a multiplex to

distinguish mealybug species found in North American vineyards. Environ. Entomol.

40: 1595-1603.

Daane, K. M., R. P. P. Almeida, V. A. Bell, M. Botton, M. Fallahzadeh, M. Mani, J.

L. Miano, R. Sforza, V. M. Walton, and T. Zaveizo. 2012. Biology and

management of mealybugs in vineyards, pp. 271-308. In N. J. Bostanian, R. Isaacs,

and C. Vincent (eds.) Arthropod Management in Vineyards. Springer, the

Netherlands.

Doutt, R. L., and K. S. Hagen. 1950. Biological control measures applied against

Pseudococcus maritimus on pears. J. Econ. Entomol. 43: 94-96.

Figadère, B. A., J. S. McElfresh, D. Borchardt, K. M. Daane, W. Bentley, and J. G.

Millar. 2007. trans-α-Necrodyl isobutyrate, the sex pheromone of the grape

mealybug, Pseudococcus maritimus. Tetrahedron Lett. 48: 8434-8437.

43 Flaherty, D. L., W. L. Peacock, L. Bettiga, and G. M. Leavitt. 1982. Chemicals losing

effect against grape mealybug. Calif. Agric. 36: 15-16.

Franco, J. C., P. Suma, E. B. da Silva, D. Blumberg, and Z. Mendel. 2004.

Management strategies of mealybug pests of citrus in Mediterranean countries.

Phytoparasitica 32: 507-522.

Frick, K. E. 1952. The value of some organic phosphate insecticides in control of grape

mealybug. J. Econ. Entomol. 45: 340-341.

Geiger, C. A., and K. M. Daane. 2001. Seasonal movement and distribution of the grape

mealybug (Homoptera: Pseudococcidae); developing a sampling program for San

Joaquin Valley vineyards. J. Econ. Entomol. 94: 291-301.

Geiger, C. A., K. M. Daane, W. J. Bentley, G. Y. Yokota, and L. A. Martin. 2001.

Sampling program for grape mealybugs improves pest management. Calif. Agric. 55:

19-27.

Golino, D. A., S. T. Sim, R. Gill, and A. Rowhani. 2002. California Mealybugs can

spread grapevine leafroll disease. Calif. Agric. 56: 196-201.

Grasswitz, T. R., and D. G. James. 2008. Movement of grape mealybug, Pseudococcus

maritimus, on and between host plants. Entomol. Exp. Appl. 129: 268-275.

44 Grimes, E. W., and W. W. Cone. 1985a. Control of the grape mealybug, Pseudococcus

maritimus, (Hom.: Pseudococcidae), on Concord grape in Washington. J. Entomol.

Soc. Brit. Columbia 82: 3-6.

Grimes, E. W., and W. W. Cone. 1985b. Life history, sex attraction, mating, and

natural enemies of the grape mealybug, Pseudococcus maritimus (Homoptera:

Pseudococcidae). Ann. Entomol. Soc. Am. 78: 554-558.

Hinkens, D. M., J. S. McElfresh, and J. G. Millar. 2001. Identification and synthesis of

the sex attractant pheromone of the vine mealybug, Planococcus ficus. Tetrahedron

Lett. 42:1619-1621.

Ifoulis, A. A., and M. Savopoulou-Soultani. 2006. Developing optimum sample size

and multistage sampling plans for Lobesia botrana (Lepidoptera: Tortricidae) larval

infestation and injury in northern Greece. J. Econ. Entomol. 99: 1890-1898.

Iwao, S. 1968. A new regression method for analyzing the aggregation of animal

populations. Res. Pop. Ecol. 10: 1-20.

Karandinos, M. G. 1976. Optimum sample size and comments on some published

formulae. Bull. Entomol. Soc. Am. 22: 417-421.

45 Madsen, H. F., and L. B. McNelly. 1960. Control of the grape mealybug on apricots. J.

Econ. Entomol. 53: 354-357.

Mallampalli, N., and R. Isaacs. 2002. Distribution of egg and larval populations of

cranberry fruitworm (Lepidoptera: Pyralidae) and cherry fruitworm (Lepidoptera:

Tortricidae) in highbush blueberries. Environ. Entomol. 31: 852-858.

Martelli, G. P., A. A. Agranovsky, M. Bar-Joseph, D. Boscia, T. Candresse, R. H. A.

Coutts, V. V. Dolja, B. W. Falk, D. Gonsalves, W. Jelkmann, A. V. Karasev, A.

Minafra, S. Namba, H. J. Vetten, G. C. Wisler, and N. Yoshikawa. 2002. The

family Closteroviridae revised. Arch. Virol. 147: 2039-2044.

Millar, J. G., and S. L. Midland. 2007. Synthesis of the sex pheromone of the obscure

mealybug, the first example of a new class of monoterpenoids. Tetrahedron Lett.

48:6377-6379.

Millar, J. G., K. M. Daane, J. S. McElfresh, J. Moreira, R. Malakar-Kuenen, M.

Guillen, and W. J. Bentley. 2002. Development and optimization of methods for

using sex pheromone for monitoring the mealybug Planococcus ficus (Homoptera:

Pseudococcidae) in California vineyards. J. Econ. Entomol. 95: 706-714.

46 Millar, J. G., J. A. Moreira, J. S. McElfresh, K. M. Daane, and A. S. Freund. 2009.

Sex pheromone of the longtailed mealybug: A new class of monoterpene structure.

Org. Lett. 11: 2683-2685.

Rayapati, N. A., S. O’Neal, D. Walsh. 2008. Grapevine leafroll disease. Washington

State University Extension Bulletin EB2027E.

Suckling, D. M. 2000. Issues affecting the use of pheromones and other semiochemicals

in orchards. Crop Prot. 19: 677-683.

Systat Software, Inc. Systat 12 2007. Systat Software, Inc.. San Jose, CA.

Taylor, L. R. 1961. Aggregation, variance and the mean. Nature 189: 732-735.

Tsai, C., A. Rowhani, D. A. Golino, K. M. Daane, and R. P. P. Almeida. 2010.

Mealybug Transmission of Grapevine Leafroll Viruses: An Analysis of Virus-Vector

Specificity. Phytopathology. 100(8): 830-834.

Walton, V. M., K. M. Daane, and K. L. Pringle. 2004. Utilizing the sex pheromone of

Planococcus ficus to improve pest management in South African vineyards. Crop

Prot. 23: 1089-1096.

47 Walton, V. M., K. M. Daane, W. J. Bentley, J. G. Millar, T. E. Larsen, and R.

Malakar-Kuenen. 2006. Pheromone-based mating disruption of Planococcus ficus

(Hemiptera: Pseudococcidae) in California vineyards. J. Econ. Entomol. 99: 1280-

1290.

Waterworth, R. A., R. A. Redak, and J. G. Millar. 2011a. Pheromone-baited traps for

assessment of seasonal activity and population densities of mealybug species

(Hemiptera: Pseudococcidae) in nurseries producing ornamental plants. J. Econ.

Entomol. 104: 555-565.

Waterworth, R. A., I. M., Wright, and J. G. Millar. 2011b. Reproductive biology of

three cosmopolitan mealybug (Hemiptera: Pseudococcidae) species, Pseudococcus

longispinus, Pseudococcus viburni, and Planococcus ficus. Ann. Entomol. Soc. Am.

104: 249-260.

Way, M. J., and H. F. van Emden. 2000. Integrated pest management in practice -

pathways towards successful application. Crop Prot. 19: 81-103.

WSU AgWeatherNet. 2011. The Washington Agricultural Weather Network Version

2.0, Washington State University, Prosser, WA, (accessed

15 February 2012).

48 Young, L. J., and J. H. Young. 1998. Statistical ecology: a population perspective.

Kluwer Academic Publishers, Norwell, MA.

Zou, Y., K. M. Daane, W. J. Bentley, and J. G. Millar. 2010. Synthesis and bioassay

of racemic and chiral trans-α-necrodyl isobutyrate, the sex pheromone of the grape

mealybug Pseudococcus maritimus. J. Ag. Food Chem. 58:4977-4982.

49 CHAPTER 3

PSEUDOCOCCUS MARITIMUS (HEMIPTERA: PSEUDOCOCCIDAE) AND

PARTHENOLECANIUM CORNI (HEMIPTERA: COCCIDAE) ARE CAPABLE

OF TRANSMITTING GRAPEVINE LEAFROLL-ASSOCIATED VIRUS 3

BETWEEN VITIS x LABRUSCANA L. AND VITIS VINIFERA.

Introduction

The grape mealybug, Pseudococcus maritimus (Ehrhorn), is one of several mealybug species infesting grapevines (Vitis spp.) in the United States of America

(Golino et al. 2002). The species is native to North America (Ben-Dov 1995) and was originally described in 1900 (Ehrhorn 1900). This species has been recorded as an economic pest on grapevines (Flebut 1922), nursery stock of Taxus spp. (Neiswander

1949), apricots (Madsen and McNelly 1960), and pears (Madsen and Westigard 1962).

However, these records are questionable due to the lack of reliable taxonomic means to differentiate the grape mealybug from the obscure mealybug (Pseudococcus viburni).

The development of more reliable taxonomic descriptors (Miller et al. 1984) and molecular methods (Daane et al 2011) have improved accurate identification of different mealybug species in the vineyards (Daane et al., 2011). The European fruit lecanium scale (Parthenolecanium corni Bouché), is a widespread species of soft scale that is known from most of Europe, Turkey, Lebanon, Russia, Algeria, Libya, and New Zealand

(Santas 1985). This soft scale insect pest is also widely distributed in the United States

(Gill 1988) with records from California, Conneticut, Florida, Illinois, Indiana, Iowa,

Kansas, Maine, Massachusetts, New York, Ohio, Rhode Island, and Tennessee (ScaleNet

50 2012). It is a polyphagous insect (Ben Dov 1993) and is typically monovoltine (Gill

1988, Sforza 2000).

Historically, it has been the indirect damage from the feeding activites of Pa. corni and Ps. maritimus has caused the greatest perceived economic injury in vineyards, due to accumulation of honeydew and subsequently growth of sooty mold on berry clusters (Gill 1988, Geiger and Daane 2001). However, it was later discovered that Ps.

maritimus is a competent vector for Grapevine leafroll-associated 3 (GLRaV-3) (Golino et al. 2002). GLRaV-3 is the most widespread virus species present in grapevines affected by grapevine leafroll disease (GLRD). In addition, a more recent study also demonstrated that Pa. corni is capable of transmitting GLRaV-1 (Hommay et al. 2008), another virus species associated with GLRD. Thus, the economic abundance thresholds of both Ps. maritimus and Pa. corni have been significantly reduced due to their status as vectors of GLRaVs. Such low abundance thresholds for both Ps. maritimus and Pa. corni have highlighted the need for faster and more reliable means for their detection (i.e. pheromone-based monitoring) within the vineyards (Bahder et al. 2013).

GLRD is the one of the most devastating viral diseases of wine grapes (Vitis vinifera) in all major grape-growing regions of the world and can cause significant yield losses (Rayapati et al. 2008). The disease is prevalent in Washington State vineyards

(Martin et al. 2005) and its symptoms in red-fruited wine grape cultivars include reddening of interveinal areas and downward rolling of leaf margins. Leaves of affected white-fruited wine grape cultivars may show chlorosis and downward rolling of leaf margins. The detection of GLRaV-3, and other GLRaVs, in V. x labruscana L. is difficult due to latent infections, rendering infected vines asymptomatic. Unlike wine grapes,

51 yield losses are not documented in V. x labruscana L. vines affected by GLRD. As a result of the perceived lack of economic impact of GLRD on V. x labruscana L., very

little attention has been paid to monitoring spread of GLRaVs or control of Ps. maritimus

and Pa. corni in Concord juice grape vineyard blocks inWashington State. On the other

hand, prophylactic systemic applications of insecticides are conventional practices taken

to minimize potential losses due to GLRD in V. vinifera vineyards. Both Pa. corni and

Ps. maritimus are primary candidates for studying the epidemiology of GLRD in

Washington State because both species have established populations in V. vinifera and V.

x labruscana L. vineyards.

While it has been demonstrated that Ps. maritimus is a competent vector of

GLRaV-3 (Golino et al. 2002) nothing is known about its role in the epidemiology of

GLRD within Concord juice grape (V. x labruscana L.) vineyards. In Washington State,

blocks of V. x labruscana L. are often grown in close proximity to V. vinifera blocks within the same vineyard. Recently, GLRaV-3 was documented for the first time from

samples derived from asymptomatic V. x labruscana L. grapevines, thus indicating that

this virus can infect other Vitis species. In addition, populations of Ps. maritimus are

known to infest V. x labruscana L. vineyards in Washington State (Frick 1952). Taken

together, it is plausible to assume that V. x labruscana L. may be playing a role in the

epidemiology of GLRD by serving as source of virus inoculum for spread by a competent

vector into adjacent V. vinifera vineyards. Thus, investigating the vectoring ability of Ps.

maritimus to transmit GLRaVs from V. vinifera to V. x labruscana L., and vice versa

would help in the understanding of the epidemiology of GLRD. In addition, although Pa.

corni has been demonstrated to transmit GLRaV-1 between wine grapes under

52 greenhouse conditions (Hommay et al. 2008), there is a lack of information on its role in

the spread of GLRaVs between V. x labrsucana L. and V. vinifera, and vice versa.

The primary objective of this research is to determine if Ps. maritimus is capable

to transmitting GLRaV-3 between V. x labruscana L. and V. vinifera grapevines in a

greenhouse environment. The secondary objective is to determine if Pa. corni is a competent vector of GLRaV-3 and can also transmit the virus between V. x labruscana L. and V. vinifera grapevines. The results of this study will increase our understanding of the role of Pa corni and Ps. maritimus in the epidemiology of GLRD in Washington State, and by extension, other major cool climate grape-growing regions of the world. This information could also assist growers in combating the spread of GLRD and its vector.

Materials and Methods

Insect Colonies and Plant Material. Specimens of Pa. corni and Ps. maritimus

were obtained as eggs from V. x labruscana L. vineyards in Prosser, WA and reared on

V. vinifera cv. Pixie (a dwarf Pinot Meunier) seedlings in a controlled climate greenhouse

at 25° - 30° C with a 14-hour day length and a 10-hour dark period. Colonies were

housed in separate Collapsible Observation and Rearing Cages (Bioquip Rancho

Dominguez, CA). Virus-free V. x labsruscana L. plantlets were initially obtained from a

commercial nursery (Inland Desert INC., Benton City, WA) and mist-propagated, using

DIP‘N GROW® (1.0% Indole-3-butyric acid, 0.5% 1-Napthaleneacetic acid) (DIP’N

GROW, INC Clackamas, OR) to stimulate root growth. Certified virus-free cuttings of

V. vinifera cv. Cabernet Sauvignon were obtained from the National Clean Plant Network

located Washington State University IAREC (Prosser, WA).

53 Identification of Pa. corni and Ps. maritimus. Species identifications were made by Dr. Gillian Watson at the Plant Pest Diagnostic Branch of the California

Department of Agriculture (Sacramento, CA). Voucher specimens of Pa. corni and Ps.

maritimus from Washington State were taken from established colonies and stored at

Washington State University’s Irrigated Agriculture Research and Extension Center

(Prosser, WA). Sequence data were obtained from the COI region of both species using primers designed by Daane et al. (2011) and Deng et al. (2012) (Table 1) to complement the morphological data. Due to the difficulty in using morphological characters for species level identification in the Coccoidea, a result of high levels of morphological variation even within species, sequence data were analyzed to ensure species identity. For this purpose, genomic DNA was extracted by grinding individual, adult insects in 25μl of

GES buffer (0.1M glycine, 50mM NaCl, 1mM EDTA, 0.5% Triton X-100) containing a

1% mercaptoethanol. The homogenate was denatured at 95° C for 10 minutes, cooled on ice for five minutes, and 5µl aliquot of the denatured extract used as template for PCR

(Rowhani et al. 2000). All reactions were a total volume of 12.5 μl, using 1μl of genomic

DNA. Concentrations for reagents were as follows; 4mM MgCl2, 0.25 μM for each

primer, 0.1 mM of each dNTP per microliter, and 0.75 U Taq DNA polymerase (Roche

Diagnostics, Mannheim, Germany) (Rowhani et al. 2000). Amplification was performed

in a GeneAmp PCR System 9700 thermal cycler (Applied Biosystems Carlsbad, CA)

with the following conditions: initial denaturation at 94° C for 30 s followed by 35 cycles

of melting at 94° C for 30 s, annealing at 55° C for 45 s, extension at 72° C for 30 s, and

a final extension at 72° C for seven minutes. To visualize amplification, five microliters

of PCR product was run on 1.5% agarose gel stained with GelRed (Biotium; Hayward,

54 Table 3-1. Primer sequences for virus and insect verification

Species Gene Direction Sequence (5’Æ3’) PCR Product Size (bp)

Pa. corni COI Forward CAGGAATAATAGGAACATCAATAAG1 550

Reverse ATCAATGTCTAATCCGATAGTAAATA1

Ps. maritimus COI Forward CTGATTTCCTTTATTAATTAATTCAAC2 400

Reverse CAATGCATATTATTCTGCCATATTA2

GLRaV-3 Hsp70 Forward CGCTAGGGCTGTGGAAGTATT3 550

Reverse GTTGTCCCGGGTACCAGT TAT3

1Deng et al. 2012

2Daane et al. 2011

3Osman and Rowhani 2006

55 CA). Photographs of gels were taken using a UV-transilluminator Biorad Universal

Hood (Bio-Rad Laboratories; Hercules, CA). Samples that produced strong bands were sequenced directly from the PCR product while those that produced weak bands were first cloned using TOPO TA Cloning® kit (Invitrogen; Carlsbad, CA) prior to sequencing.

Transmission Bioassays. A total of 90 transmission bioassays were conducted from 2011 to 2012. Three different experiments were conducted for both species of insect, Pa. corni and Ps. maritimus. For both species, acquisition and inoculation access periods of one week each were tested, where 10 first instar individuals were allowed to feed on the virus source plant for a one-week acquisition access period (AAP) and were then transferred to a virus-f ree re cipient plant for a one-week inoculation access period

(IAP). This wa s acco mplis hed b y cutti ng the section of the leaf where the insects were feeding and pla cing th e lea f section on the recipient plant, allowing the insect to naturally change hosts and avoiding direct contact with the insect so as to reduce mortality. The transfer o f 10 in divid uals fr om so urce t o recipient was replicated 15 times across all experime nts. Tr ansm ission bioassays consisted of transferring individuals from an infected V. vini fera plant to virus free V. x labruscana L. plants, an infected V. x labruscana L. plant to virus free V. x labruscana L., and from an infected V. x labruscan a L. to viru s free V. vin ifera plants . Plants were then tested at two week, one month, th ree m onth, a nd six -mon th int ervals post-IAP. Specimens of both Pa. corni and

Ps. marit imus were collected from the GLRaV-3-positive sour ce plant and extract from these inse cts we re tes ted by RT-P CR to verif y that both insects could acquire the virus.

As a positive control to validate the transmission p rotoco l, fiv e transmissions for each species of vecto r was conducted betwe en plants of V. vin ifera using the same technique

56 used for the interspecific transmission assays since both insect species have previously been shown to transmit GLRaVs between plants of V. vinifera (Golino et al 2002; Sforza et al. 2003; Tsai et al. 2010; Hommay et al. 2008). To test for the presence or absence of

GLRaV-3 in recipient plants, the positive source plant, and in each vector species, single- tube RT-PCR reactions were performed using primers targeting a portion of the heat- shock protein-70 homolog (HSP70h) gene of GLRaV-3 (Table 1). Extraction of virus genetic material from plants was accomplished by grinding 25 mg of petiole tissue in 5 ml of grape extraction buffer (GEB; 1.59 g/L Na2CO3, 2.93 g/L NaHCO3, 2% polyvinyl pyrrolidone, 0.2% bovine serum albumin, 0.5% Tween 20) and the extracts were stored in -80° C until use. To extract viral RNA, 4 μl of the GEB extract was denatured in 25 μl of GES solution (containing 1% mercaptoethanol) as described above. All RT-PCR reactions, reagent concentrations, and visualization steps for verifying the presence or absence of GLRaV-3 are the same as described above.

Data Analysis. Sequence data were analyzed using Vector NTI (Life

Technologies, Grand Island, NY) and MEGA 5 (Tamura et al. 2011). Logit analysis was used to determine whether probability of infection was affected by vector (Pa. corni vs.

Ps. maritimus) and by direction of the transmission (V. vinifera to V. x labruscana L., V. x labruscana L. to V. x labruscana L., and V. x labruscana L. to V. vinifera). The analysis was done using PROC GENMOD in SAS (SAS Institute 2010). The saturated model with main effects and the interaction term was fitted. The ilink option was used to transform means back to proportions. The odds ratio for the vector effect was calculated using the estimate option.

57 Results

Species Identity. In pair-wise comparisons, sequence data for the COI region of

Ps. maritimus from the colony used in the transmission bioassays (KC679784) shared

99% nucleotide identity with Ps. maritimus sequence data obtained from GeneBank

(JN112800) and 91% nucleotide identity with Ps. viburni (JN112803) (Figure 1). Also, sequence data for the COI region of Pa. corni from the colony used in the transmission bioassays (KC679783) shared 99% nucleotide identity with Pa. corni sequence data obtained from GeneBank (JQ795617)(Figure 2). Lastly, Hsp70h-derived sequences of the

GLRaV-3 isolate (KC113195) obtained from the virus-infected source plant shared 99% nucleotide identity with the WA isolate (DQ780890)(Figure 3). Gene fragments of COI for Pa. corni were 399 bps (Figure 1), gene fragments of COI for Ps. maritimus were 543 bps (Figure 2), and gene fragments of hsp-70 from GLRaV-3 were 546 bps (Figure3).

Based on nucleotide identity and fragment sizes produced, these results confirmed the identities of the virus and vector species used in this study.

Transmission Bioassays. Transmission success of Ps. maritimus varied between

26.7% and 53.3%, whereas success of Pa. corni was substantially lower at 6.7%. We observed transmission in all three directions. The logit analysis revealed a highly significant vector effect (df =1, Chi-Square=12.10, P=0.0005.). The odds ratio for the vector effect was 7.1 (confidence interval of 2.0 to 24.4), indicating that the likelihood of infection under controlled greenhouse conditions was about sevenfold higher if Ps. maritimus was the vector than if Pa. corni was the vector. Transmission success was not affected by direction of the transmission (df=2, Chi-Square=0.11, P=0.9453) among grape species. The interaction between vector and transmission direction among grape

Figure 3-1. Sequence data for the COI gene of Ps. maritimus; Pmar_COI_Daane

represents sequence data for Ps. maritimus provided by Daane et al. 2011, Pmar_COI

represents COI sequence data obtained from individuals used in transmission bioassays,

and Pvib_COI represent COI sequence data from Ps. viburni, the sister species to Ps.

maritimus.

species also was also not significant (df=2, Chi-Square=1.81, P=0.4041). Of the individuals of Pa. corni that were tested for presence of GLRaV-3 from the positive

source plant, 41% tested positive for GLRaV-3, whereas 17% of Ps. maritimus sampled

tested positive for GLRaV-3. For the positive control bioassays, there was 20%

successful transmission for Pa. corni and 40% successful transmission for Ps. maritimus.

Both species of vector were capable of transmitting GLRaV-3 from V. vinifera to

V. x labruscana L. under experimental conditions. Of the 15 replicates performed for

each vector species, there were six successful transmissions for Ps. maritimus (Figure 4),

whereas Pa. corni had one successful transmission (Figure 5). Both species of vector

Figure 3-2. Sequence data for the COI gene of Pa. corni; Pcor_COI_Deng represents

sequence data for Pa. corni provided by Deng et al. (JQ795617.1) and Pcor_COI represents sequence data for Pa. corni used in transmission bioassays.

were capable of transmitting GLRaV-3 from V. x labruscana L. to V. x labruscana L. under experimental conditions. Of the 15 replicates performed for each vector species, there were four successful transmissions for Ps. maritimus (Figure 6), whereas Pa. corni had one successful transmission (Figure 7). Both species of vector were capable of transmitting GLRaV-3 from V. x labruscana L. to V. vinifera under experimental conditions. Of the 15 replicates performed for each vector species, there were eight successful transmissions for Ps. maritimus (Figure 8), whereas Pa. corni had one

successful transmission (Figure 9).

Discussion

Based on the results of this study, it can be concluded that both Pa. corni and Ps.

maritimus are competent vectors of GLRaV-3 between V. x labruscana L. and V. vinifera

Figure 3-3. Sequence data for hsp-70 gene of GLRaV-3 used in transmission bioassays;

sequence data obtained from virus in source plant (1), sequence data obtained from virus

within the insect vector (2), and sequence data obtained from virus in recipient Concord

plant (3).

under controlled greenhouse conditions. While Pa. corni appears to be more efficient at

acquiring GLRaV-3, Ps. maritimus seems to be more efficient at transmitting the virus under controlled greenhouse conditions. The differences in the abilities of these two

species to acquire and transmit GLRaV-3 may be a result of the biology of each insect or

the result of stress from changing host plants. Ps. maritimus is typically more mobile;

migrating to different regions of the vine throughout the season and among generations,

whereas Pa. corni, are typically sedentary once a suitable feeding site is found. This

Figure 3-4. Gel from RT-PCR test of recipient V. x labruscana L. plants exposed to first instars of Ps. maritimus that had been exposed to the virus infected V. vinifera source; C

= Concord and number represents plant number.

Figure 3-5. Gel from RT-PCR test of recipient V. x labruscana L. plants exposed to first instars of Pa. corni that had been exposed to the virus infected V. vinifera source; C =

Concord and number represents plant number.

Figure 3-6. Gel from RT-PCR test of recipient V. x labruscana L. plants exposed to first instars of Ps. maritimus that had been exposed to the virus infected V. x labruscana L. source; C = Concord and number represents plant number.

62 behavior may have caused Pa. corni to be more prone to stress, translating into lower

survival or transmission rate when they were moved to a new food source in these

studies. While Pa. corni and Ps. maritimus were shown to be competent vectors of

GLRaV-3 between V. x labruscana L. and V. vinifera under experimental conditions, it is

not known if this transmission occurs in a natural, vineyard setting and if it does, if it is

economically significant. There are variables present in a natural setting, such as

presence of insect predators and parasitoids, adverse weather conditions, and pesticides

that may prevent successful, frequent transmission or movement of 1st instar crawlers

among vineyards. In this light, future research will be needed to evaluate the economic

impact of spread of GLRaV-3 by mealybugs and scale insects between V. x labrsucana

L. and V. vinifera. However, even if the movement of crawlers between vineyard types is

very low and the transmission rate of GLRaVs in the field is very low, the potential exists

that insect transmission of GLRaV-3 can still pose a threat to sustainability of the grape

industry in the long term.

Due to the impact of GLRD on the wine grape industry, a more collaborative

effort may be required between growers of V. x labruscana L. and V. vinifera to more

effectively manage the spread of this disease in Washington State. One major aspect in the prevention of the spread of GLRD is control of vectors. In Washington State, this is accomplished through chemigation, a technique where a systemic insecticide is introduced through a drip irrigation system. Chemigation is the preferable means to control Ps. maritimus in vineyards because it does not disrupt populations of natural

enemies that are effective at keeping Ps. maritimus populations low and that it targets all

feeding instars. While successful in V. vinifera vineyards, this technique is not applicable

Figure 3-7. Gel from RT-PCR test of recipient V. x labruscana L. plants exposed to first instars of Pa. corni that had been exposed to the virus infected V. x labruscana L. source;

C = Concord and number represents plant number.

Figure 3-8. Gel from RT-PCR test of recipient V. vinifera plants exposed to first instars

of Ps. maritimus that had been exposed to the virus infected V. x labruscana L. source;

W = Wine grape and number represents plant number.

Figure 3-9. Gel from RT-PCR test of recipient V. vinifera plants exposed to first instars

of Pa. corni that had been exposed to the virus infected V. x labruscana L. source; W =

Wine grape and number represents plant number.

64 to V. x labruscana L. in Washington State because the majority of vineyards of V. x labruscana L. are irrigated by overhead sprinklers, necessitating traditional spray insecticides to control pest populations. Thus, vineyards of V. x labruscana L. could be harboring populations of Ps. maritimus that could transmit GLRD to neighboring V. vinifera vineyards. Alternatively, the spray regimens being applied in V. x labruscana L. vineyards may have a negative impact on populations of natural enemies of Ps. maritimus and Pa. corni, leading to a build up of these pests in neighboring vineyards.

The results obtained in this study demonstrate for the first time that interspecies transmission of GLRaVs in grapes, V. x labruscana L. and V. vinifera, by Pa. corni and

Ps. maritimus is possible under controlled greenhouse conditions in Washington State.

While the degree, to which this is occurring under natural conditions is unknown, the risk that GLRD poses to the sustainability of the wine industry highlights the need for immediate steps in expanding IPM programs to include V. x labruscana L. Since the biological possibility exists for GLRaV-3 to be transmitted from V. x labruscana L. to V. vinifera under experimental conditions, it is likely that this transmission is occurring under field conditions. Because of this risk, future work should focus on evaluating the frequency at which GLRaV-3 is moving among vineyards of V. x labruscana L. and V. vinifera as well as the economic and biological significance of GLRaV-3 moving between vineyards of V. x labruscana L. and V. vinifera so that responsible management decisions can be made that will allow for the sustainability of the wine grape industry in

Washington State.

65 References

Bahder, B. W., R. A. Naidu, K. M. Daane, J. G. Millar, and D. B. Walsh. 2013.

Pheromone-Based Monitoring of Pseudococcus maritimus (Hemiptera:

Pseudococcidae) Populations in Concord Grape Vineyards. Journal of Economic

Entomology.

Ben-Dov, Y. 1995. A systematic catalogue of the mealybugs of the world (Insecta:

Homoptera: Coccoidea: Pseudococcidae, and Putoidae) with data on geographical

distribution, host plants, biology, and economic importance. Intercept, Hampshire,

UK.

Belli, G., A. Fortusini, P. Casati, L. Belli, P. A. Bianco, and S. Prati. 1994.

Transmission of a grapevine leafroll associated closterovirus by the scale insect

Pulvinaria vitis L. Rivista di Patoliga Vegetale. S. V. 4: 105-108.

Cabaleiro, C. and A. Segura. 1997. Field Transmission of Grapevine Leafroll

Associated Virus 3 (GLRaV-3) by the Mealybug Planococcus citri. Plant

Disease. 81: 283-287.

66 Daane, K. M., M. C. Middleton, R. Sforza, M. L. Cooper, V. M. Walton, D. B.

Walsh, T. Zaviezo, and R. P. P. Almeida. 2011. Development of a Multiplex

PCR for Identification of Vineyards Mealybugs. Environmental Entomology.

40(6): 1595-1603.

Deng, J., F. Yu, T. Zhang, H. Hu, C. Zhu, S. Wu, and Y. Zhang. 2012. DNA

Barcoding of six Ceroplastes species (Hemiptera: Coccoidea: Coccidae) from

China. Molecular Ecology Resources. 12: 791-796.

Ehrhorn, E. M. 1900. New Coccidae From California. The Canadian Entomologist. 32:

311-318.

Flebut, A. J. 1922. The Grape Mealybug. California Department of Agriculture Bulletin.

11: 7-11.

Frick, K. E. 1952. The Value of Some Organic Phosphate Insecticides in Control of

Grape Mealybug. Journal of Economic Entomology. 45(2): 340-341.

Fortusini, A., G. Scattini, S. Prati, S. Cinquanta, and G. Belli. 1997. Transmission of

grapevine leafroll virus 1 (GLRaV-1) and grapevine virus A (GVA) by scale

insects. Proceedings of the International Council of the Study of Viruses and

Virus-like Disease of Grapevines. Pages 121-122.

Geiger, C. A., and K. M. Daane. 2001. Seasonal Movement and Distribution of the

Grape Mealybug (Homoptera: Pseudococcidae): Developing a Sampling Program

for San Joaquin Valley Vineyards. Journal of Economic Entomology. 94(1): 291-

301.

Golino, D. A., S. T. Sim, R. Gill, and A. Rowhani. 2002. California mealybugs can

spread grapevine leafroll disease. California Agriculture. 56(6): 196-201.

Hommay, G., V. Komar, O. Lemaire, and E. Herrbach. 2008. Grapevine virus A

transmission by larvae of Parthenolecanium corni. European Journal of Plant

Pathology. 121: 185-188.

Madsen, H. F., and L. B. McNelly. 1960. Control of the grape mealybug on apricots.

Journal of Economic Entomology. 53: 354-357.

Madsen, H. F., and P. H. Westigard. 1962. Behavior and control of the grape mealybug

on pear. Journal of Economic Entomology. 55: 849-850.

Mahfoudhi, N., M. Digiaro, and M. H. Dhouibi. 2009. Transmission of Grapevine

Leafroll Viruses by Planococcus ficus (Hemiptera: Pseudococcidae) and

Ceroplastes rusci. Plant Disease. 93: 999-1002.

68 Martin, R. R., K. C. Eastwell, A. Wagner, S. Lamprecht, and I. E. Tzanetaki. 2005.

Survey for Viruses of Grapevine in Oregon and Washington. Plant Disease.

89(7): 763-766.

Miller, D. R., R. J. Gill., and D. J. Williams. 1984. Taxonomic analysis of

Pseudococcus affinis (Maskell), a senior synonym of Pseudococcus obscures

Essig, and a comparison with Pseudococcus maritimus (Ehrhorn) (Homoptera:

Coccoidea: Pseudococcidae).

Naidu, R. A., S. O’Neal, and D. B. Walsh. 2008. Grapevine leafroll disease.

Washington State University Bulletin EB2027E.

Neiswander, R. B. 1949. The grape mealybug on Taxus in Ohio. Journal of Economic

Entomology. 42: 41-43.

Osman, F. and A. Rowhani. 2006. Application of a spotting sample preparation

technique for the detection of pathogens in woody plants by RT-PCR and real-

time PCR (TaqMan). Journal of Virological Methods. 133: 130-136.

Petersen, C. L. and J. G. Charles. 1997. Transmission of grapevine leafroll-associated

closteroviruses by Pseudococcus longispinus and P. calceolariae. Plant

Pathology. 46: 509-515.

69 Rowhani, A., L. Biardi, R. Johnson, P. Saldarelli, Y. P. Zhang, J. Chin, and M.

Green. 2000. Simplified sample preparation method and one-tube RT-PCR for

grapevine viruses. 13th ICGV, Extended Abstracts, pp148.

Santas, L. A. 1985. Parthenolecanium corni (Bouché) and Orchard Scale Pest Producing

Honeydew Foraged by Bees in Greece. Entomologia Hellenica. 3(2): 53-58.

Sforza, R., E. Boudon-Padieu, and Charles Greif. 2003. New mealybug species

vectoring Grapevine leafroll-associated viruses-1 and -3 (GLRaV-1 and -3).

European Journal of Plant Pathology. 109: 975-981.

Sim, S. T., A. Rowhani, and D. A. Golino. 2003. Experimental transmission of

grapevine leafroll-associated virus 5 and 9 by longtailed mealybug. Proceedings

of the International Council of the Study of Viruses and Virus-like Disease of

Grapevines. Pages 211-212.

Soule, M. J., K. C. Eastwell, and R. A. Naidu. 2006. First Report of Grapevine leafroll

associated virus-3 in American Vitis spp. Grapevines in Washington State. Plant

Disease. 90(11): 1461.

Tsai, C. W., J. Chau, L. Fernandez, D. Bosco, K. M. Daane, and R. P. P. Almeida.

2008. Transmission of Grapevine leafroll-associated virus-3 by the Vine

Mealybug (Planococcus ficus). Phytopathology. 98(10): 1093-1098.

70 Tsai, C. W., A. Rowhani, D. A. Golino, K. M. Daane, and R. P. P. Almeida. 2010.

Mealybug Transmission of Grapevine Leafroll Viruses: An Analysis of Virus-

Vector Specificity. Phytopathology. 100: 830-834.

Tamura, K., D. Peterson, N. Peterson, G. Stecher, M. Nei, and S. Kumar. 2011.

MEGA5: Molecular Evolutionary Genetics Analysis using Maximum Likelihood,

Evolutionary Distance, and Maximum Parsimony Methods. Molecular Biology

and Evolution. 28:2731-2739.