Citation
Citrus tristeza virus

Material Information

Title:
Citrus tristeza virus cross-protection, graft transmission, and serology
Creator:
Rocha-Pena, Mario Alberto, 1950-
Publication Date:
Language:
English
Physical Description:
xii, 104 leaves : ill., photos (some col.) ; 29 cm.

Subjects

Subjects / Keywords:
Antibodies ( jstor )
Antigens ( jstor )
Bark ( jstor )
Enzyme linked immunosorbent assay ( jstor )
Inoculum ( jstor )
Monoclonal antibodies ( jstor )
Orange fruits ( jstor )
Plants ( jstor )
Receptors ( jstor )
Sour oranges ( jstor )
Dissertations, Academic -- Plant Pathology -- UF
Plant Pathology thesis Ph. D
Genre:
bibliography ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1990.
Bibliography:
Includes bibliographical references (leaves 94-103).
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Mario Alberto Rocha-Pena.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. This item may be protected by copyright but is made available here under a claim of fair use (17 U.S.C. §107) for non-profit research and educational purposes. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact the RDS coordinator (ufdissertations@uflib.ufl.edu) with any additional information they can provide.
Resource Identifier:
027771659 ( ALEPH )
26498452 ( OCLC )

Downloads

This item has the following downloads:


Full Text











CITRUS TRISTEZA VIRUS: CROSS-PROTECTION, GRAFT
TRANSMISSION, AND SEROLOGY



















By

MARIO ALBERTO ROCHA-PERA


















A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 1990






























To my beloved wife Alicia, and to my adored son Eric, whom are the world of my life.



To my parents Herminio and Dolores, and my brothers and sisters.








ACKNOWLEDGEMENTS


I want to express my sincere gratitude and acknowledgment to the following persons, institutions, and organizations, who supported my graduate studies at the University of Florida.

The financial support from Consejo Nacional de Ciencia y Tecnologia (CONACyT), as well as the support and leave of absence from Instituto Nacional de Investigaciones Forestales y Agropecuarias (INIFAP), both institutions from M6xico, is greatly appreciated.

To my co-major professors, Drs. R.F. Lee and C.L. Niblett, for their constant encouragement, guidance and support throughout the course of the thesis work, and for their personal interest in my academic preparation. To them and the rest of the supervisory committee, Drs. S.M. Garnsey, D.E. Purcifull, and R.K. Yokomi, for valuable suggestions in revision of the manuscript.

The financial support from the Florida High Technology and Industrial Council, and the Florida Citrus Production Managers' Association, to carry out some parts of the thesis work also is acknowledged.

I thank Drs. S.M. Garnsey and T.A. Permar, USDA Orlando, and Drs. P. Moreno and M. Cambra, IVIC Valencia (Spain), for supplying the MCA-13 and 3DF1 monoclonal antibodies, respectively.




ii








The technical assistance and friendship of N. Berger, S. Marquardt, T. Nguyen, S. Jackson, and J. Zellers, as well as the help of T. Zito in the photographic work, and M. Ahnger with the statistical analysis, all at the Citrus Research and Education Center, at Lake Alfred, is greatly appreciated.

The warm friendship I found in my fellow students, lab technicians, administrative staff, and most faculty members at the Plant Pathology Department, Gainesville, and at the Citrus Research and Education Center, Lake Alfred, was indeed encouraging and will be unforgettable.

Finally, I want to thank my wife Alicia for her constant encouragement and patience to endure the hardship of my pursuit of this graduate degree.



























iii















TABLE OF CONTENTS

Paae

ACKNOWLEDGMENTS ........................ ii

LIST OF FIGURES............................................ vi

LIST OF TABLES .......... ............... ................. ix

ABSTRACT.... ............................................. xi

CHAPTER

1. INTRODUCTION....................................... 1

2. EVALUATION OF THE PROTECTING EFFECTS OF SOME
FLORIDA ISOLATES OF CITRUS TRISTEZA VIRUS AGAINST
THE DEVELOPMENT OF THE DECLINE SYNDROME ............ 6

Introduction...................................... 6
Materials and Methods .............................. 9
Virus isolates and donor hosts .............. 9
Inoculation of CTV isolates and receptor hosts. 10 Serological tests............................. 11
Detrimental effects of mild isolates and
evaluation of cross-protection................. 14
Results ............................................ 14
Antigen titers of mild and severe CTV
isolates with polyclonal and MCA-13
monoclonal antibodies........................... 14
Detrimental effect of mild isolates and
evaluation of cross-protection................ 21
Discussion ......................................... 27

3. EFFECTIVENESS OF DIFFERENT CITRUS SPECIES AS
DONOR HOSTS FOR GRAFT TRANSMISSION OF CITRUS
TRISTEZA VIRUS ...................................... 38

Introduction....................................... 38
Materials and Methods............................. 40
Virus isolates and donor hosts............... 40
Grafting procedures and receptor hosts......... 41
Virus distribution and antigen
concentration in host tissues .................. 41
Purification of CTV............................. 42

iv









Serological tests............................... 42
Results ........................................... . 43
Graft transmission of citrus tristeza
virus isolates ........................... ...... 43
Virus distribution and antigen
concentration in host tissues ................... 47
Discussion........... ................. ............ 52

4. DEVELOPMENT OF A DOT-IMMUNOBINDING ASSAY FOR
CITRUS TRISTEZA VIRUS ............................... 57

Introduction..................................... .. 57
Materials and Methods ............................. 59
Antisera used ................................. 59
Sample preparation............................. 60
Dot-immunobinding assay .................... 61
DAS-ELISA................. ..................... 62
DAS-indirect ELISA............................. 63
Evaluation ................................... 63
Evaluation of different buffers for
sample extraction ........ .................... 64
Results ............................................ 65
Development of the dot-immunobinding assay ....... 65 Evaluation..................................... 69
Evaluation of different buffers for
sample extraction .............................. 77
Discussion...................................... 77

5. SUMMARY AND CONCLUSIONS .............................. 86

LITERATURE CITED .............................. .... ... 94

BIOGRAPHICAL SKETCH ......................................... 104




















v










LIST OF FIGURES

Page

Figure 2.1 Plot of purified citrus tristeza
virus (CTV) against optical density.
Bark of healthy Citrus excelsa
(0.5 g) tissue was ground in 5.0 ml
of phosphate buffered saline, pH 7.6,
+ 0.05% Tween + 2% polyvinyl pyrrolidone and mixed with purified CTV T26 isolate
to give the desired optical density
at 260 nm (OD2@). DAS-ELISA was
performed as described in materials
and methods. An extinction
coefficient of 2.0 was assumed
(Gonsalves et al. 1978) to estimate
the relative virus concentration.......... 22

Figure 3.1 Plot of purified citrus tristeza
virus (CTV) against optical density.
Bark of healthy Citrus excelsa
(0.25 g) tissue was ground in 5.0 ml of phosphate buffered saline, pH 7.6,
+ 0.05% Tween + 2% polyvinyl pyrrolidone
and mixed with purified CTV T26 isolate
to give the desired optical density
at 260 nm (OD2). DAS-ELISA was
performed as described in materials
and methods. An extinction
coefficient of 2.0 was assumed
(Gonsalves et al. 1978) to estimate
the relative virus concentration.......... 51

Figure 4.1 Effect of different blocking solutions
on the reaction of polyclonal antibodies no. 1053 in dot-immunobinding assay with
citrus tristeza virus (CTV) isolates:
A. TBS alone; B. 3% bovine serum albumin (BSA); C. 3% gelatin; D. 0.5% non-fat dry
milk; E. 5% Triton X-100. Number 1,
CTV T66a; 2, CTV T26; 3, buffer extract
from healthy sweet orange plants;
4, TBS-Tween. Reaction conditions are
described in materials and methods........ 66

Figure 4.2 Reaction of polyclonal and monoclonal
antibodies in dot-immunobinding assay
with citrus tristeza virus (CTV) isolates.
A. Polyclonal 1053 (1.0 gg/ml) not crossvi








absorbed with buffer extract of healthy
plant. B. Polyclonal 1053 (1.0 Ag/ml)
incubated with 1:200 (v/v) buffer extract
from healthy plants for 2 hr at 370C
prior to use. C and D, monoclonal 3DF1
(1.0 Ag/ml) and MCA-13 (1:5,000 dilution)
antibodies not cross-absorbed with healthy
extract. Number 1, CTV T66a; 2, CTV T26;
3, buffer extract from healthy sweet orange plants; 4, TBS-Tween. Reaction conditions
are described in materials and methods...... 68

Figure 4.3 Relative sensitivity level of different
polyclonal and monoclonal antibodies
specific to citrus tristeza virus (CTV)
in dot-immunobinding assay. Rows A, B and C,
polyclonal antibodies nos. 1051, 1052, and
1053, respectively. Rows D and E,
monoclonal antibodies 3DF1 and MCA-13,
respectively. Extract of Citrus excelsa
greenhouse grown plants infected with
CTV T36 isolate was prepared in
TBS-Tween 1:10 (w/v) and two fold
diluted with extract of healthy plants.
Reaction conditions are described in
materials and methods ....................... 70

Figure 4.4 Reaction of polyclonal antibodies no.
1053 and 3DF1 and MCA-13 monoclonal
antibodies in dot-immunobinding assay
with twelve selected citrus tristeza
virus (CTV) isolates. Row A: 1 = Tlla;
2 = T26; 3 = T30; 4 = T50a; 5 = T55a;
6 = T3; 7 = T4; 8 = T36. Row B. 1 = T62a;
2 = T65a; 3 = T66a; 4 = T67a. Also in Row B
are extracts of healthy plants: 5 = Citrus
excelsa; 6 = Madam Vinous sweet orange;
7 = Duncan grapefruit; 8 = Mexican lime.... 74

Figure 4.5 Evaluation of different extraction
buffers on the sensitivity of DIBA with
citrus tristeza virus (CTV) isolates T26,
T62a and T66a. TBS = Tris buffered saline,
TBST = TBS containing 0.05% Tween, PBS = phosphate buffered saline, PBST = PBS +
0.05% Tween, Carb = carbonate buffer, CarbT = Carb + 0.05% Tween, HC = healthy control.
The IgG of polyclonal antibodies No. 1053 or monoclonal antibody 3DF1 were used as
primary antibodies followed by the goat antirabbit and anti- mouse IgG conjugates,
respectively. Samples were 2 gl of buffer

vii









extracts of greenhouse-grown Madam Vinous sweet orange plants infected with the CTV
isolates ground at a 1:10 dilution.......... 78

Figure 4.6 Evaluation of different extraction
buffers on the sensitivity of DAS-ELISA
and DAS-indirect ELISA with citrus tristeza
virus (CTV) isolates T26, T62a and T66a.
TBS = Tris buffered saline, TBST = TBS+Tween
(0.05%), PBS = phosphate buffered saline,
PBST = PBS+Tween (0.05%), Carb = carbonate
buffer, CarbT = Carb+Tween (0.05%),
HC = healthy control. The IgG of
polyclonal antibodies No. 1053 was used
to coat the plates for both DAS-ELISA and DAS-indirect ELISA. For DAS-ELISA the No.
1053 IgG conjugate was used as second antibody. For DAS-indirect ELISA the
unlabeled 3DF1 monoclonal antibody was the intermediate antibody followed by the goat
anti-mouse IgG conjugate. Samples were
200 Al of buffer extracts of
greenhouse-grown Madam Vinous sweet orange
plants infected with the CTV isolates
ground at a 1:10 dilution.................... 79




























viii













LIST OF TABLES

Paae

Table 2.1 Relative antigen titer of citrus
tristeza virus (CTV) mild isolates
in plants unchallenged and
challenged by the T66a severe isolate: I. Warm temperature,
Valencia/sour orange ......................... 16

Table 2.2 Relative antigen titer of citrus
tristeza virus (CTV) mild isolates
in plants unchallenged and
challenged by the T66a severe
isolate: II. Warm temperature,
Valencia/macrophylla........................ 17

Table 2.3 Relative antigen titer of citrus
tristeza virus (CTV) mild isolates
in plants unchallenged and
challenged by the T66a severe
isolate: III. Cool temperature,
Valencia/sour orange ....................... 19

Table 2.4 Relative antigen titer of citrus
tristeza virus (CTV) mild isolates
in plants unchallenged and
challenged by the T66a severe
isolate: IV. Cool temperature,
Valencia/macrophylla........................... 20

Table 2.5 Effect of citrus tristeza virus (CTV)
mild isolates on the development of the
CTV decline syndrome in plants
unchallenged and challenged by the
T66a severe isolate: I. Warm temperature,
Valencia/sour orange ........................ 23

Table 2.6 Effect of citrus tristeza virus (CTV)
mild isolates on the development of the
CTV decline syndrome in plants
unchallenged and challenged by the
T66a severe isolate: II. Warm temperature,
Valencia/macrophylla ....................... 25



ix









Table 2.7 Effect of citrus tristeza virus (CTV)
mild isolates on the development of the
CTV decline syndrome in plants
unchallenged and challenged by the
T66a severe isolate: III. Cool temperature,
Valencia/sour orange ....................... 26

Table 2.8 Effect of citrus tristeza virus (CTV)
mild isolates on the development of the
CTV decline syndrome in plants
unchallenged and challenged by the
T66a severe isolate: IV. Cool temperature,
Valencia/macropylla ....................... 28

Table 3.1 Transmission of citrus tristeza virus by
graft inoculation between selected
citrus hosts: I. Efficiency of leaf and
bark pieces as inoculum....................... 45

Table 3.2 Transmission of citrus tristeza virus by
graft inoculation between selected
citrus hosts: II. Overall rate of
transmission.................................. 46

Table 3.3 Transmission of citrus tristeza virus by
graft inoculation between selected citrus hosts: III. Effect of virus
isolates ....................................... 48

Table 3.4 Relative antigen titer of citrus
tristeza virus in different tissues of Citrus excelsa and Madam Vinous sweet
orange host plants, as measured by enzymelinked immunosorbent assay.................... 50

Table 4.1 Relative sensitivity of DIBA, DAS-ELISA and
DAS-indirect with polyclonal (1053) and monoclonal (3DF1 and MCA-13) antibodies
specific to citrus tristeza virus.............. 71

Table 4.2 Comparison of DIBA, DAS-ELISA and
DAS-indirect ELISA for the detection
of citrus tristeza virus (CTV) and
relative reactivity of polyclonal
(1053) and monoclonal (3DF1) (MCA-13)
antibodies with CTV isolates .................. 75







x














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

CITRUS TRISTEZA VIRUS: CROSS-PROTECTION, GRAFT
TRANSMISSION, AND SEROLOGY

By

MARIO ALBERTO ROCHA-PENA

DECEMBER 1990


Chairperson: Dr. Richard F. Lee Major Department: Plant Pathology

The objectives of this research were i) to evaluate some citrus tristeza virus (CTV) mild isolates under greenhouse conditions for cross protecting ability against the decline syndrome, and ii) to develop methods for detection of severe CTV challenge isolates in mixed infections in cross-protection experiments. Valencia sweet orange plants budded on sour orange rootstock were graft-inoculated by leaf pieces using any of four different mild CTV isolates and subsequently graft-challenged with a severe CTV isolate. Treatments were evaluated at temperature regimens of 21-380C and 21-33oC. Plants pre-inoculated with mild isolates when challenged with the severe isolate gave relatively lower ELISA values as compared to the unprotected, challenged control plants. The MCA-13 monoclonal antibody provided a rapid method to detect



xi








the severe isolate in mixed infections. CTV-induced decline (CTV-ID) occurred irregularly within the first 10 months after challenge inoculation at both temperature regimens. The preliminary evaluation of the cross-protecting ability of mild isolates against the CTV-ID in plants on sour orange rootstock can be accomplished under greenhouse conditions in a relatively short time of 18-24 months.

Differences were found in the effectiveness of certain tissues and/or hosts for graft-transmission of CTV. Leafpiece grafts transmitted CTV at a 90% rate vs a 75% rate using bark pieces. Madam Vinous sweet orange was the most efficient donor host giving 90% transmission to three receptor hosts, followed by Mexican lime at 85%, and Citrus excelsa at 72%. The dot-immunobinding assay (DIBA) was adapted for CTV diagnosis by using several polyclonal and monoclonal antibodies specific for CTV. The DIBA was as sensitive as DAS-ELISA and DAS-indirect ELISA for CTV detection and provides a reliable alternative for diagnosis of CTV.

















xii














CHAPTER 1
INTRODUCTION





Citrus tristeza virus (CTV) is distributed in citrus producing areas worldwide and is the most economically important viral disease of citrus (Bar-Joseph et al. 1979a, 1981, 1989). The virus infects nearly all species, varieties, and intergeneric hybrids of citrus, and some citrus relatives (Bar-Joseph et al. 1979a; Garnsey and Lee, 1988; Muller and Garnsey, 1984). However, the most destructive damage is the induced decline in scions grafted on sour orange (Citrus aurantium L.) rootstock. Some CTV isolates cause stem pitting and loss of plant vigor on some orange and grapefruit scions regardless of the rootstock (Bar-Joseph et al. 1979a, 1981, 1989).

Citrus tristeza virus is a phloem-limited, flexuous closterovirus approximately 2,000 x 11 nm in size, transmitted by aphids in a semi-persistent manner (Bar-Joseph et al. 1979a; Lister and Bar-Joseph, 1981). A single stranded positive sense RNA of 5.4-6.5 x 106 daltons has been isolated from purified virus preparations (Bar-Joseph et al. 1985). Several coat proteins of about Mr 28,000 (Guerri et al. 1990),


1








2

23,000 and 21,000 (Lee et al. 1988b), respectively, have been associated with CTV virions. The virus is readily transmitted by budding and grafting (Bennett and Costa, 1949; Bar-Joseph and Lee, 1990; Bar-Joseph et al. 1979a). Mechanical transmission has been accomplished by slash inoculation of partially purified virus preparations into the stem of hosts such as citron (Citrus medica L.) and Mexican lime {C. aurantifolia (Christm.) Swing.} (Garnsey and Muller, 1988; Garnsey et al. 1977; Muller and Garnsey, 1984). Seed transmission has not been demonstrated (McClean, 1957; Wallace, 1978).

Citrus tristeza virus occurs naturally with a diversity of isolates or strains which may differ greatly in their biological properties, such as symptomatology in different citrus hosts (Garnsey et al. 1987; McClean, 1974), aphid transmissibility (Bar-Joseph and Loebenstein, 1973; Bar-Joseph et al. 1977; Roistacher, 1981; Yokomi and Garnsey, 1987), and sensitivity to warm temperatures (Ieki and Yamada, 1980; Roistacher et al. 1974).

Several sensitive and relatively rapid methods have been developed to diagnose the presence of CTV in infected plants. These methods include SDS-immunodiffusion procedures (Garnsey et al. 1979; Bar-Joseph et al. 1980), enzyme-linked immunosorbent assay (ELISA) (Bar-Joseph et al. 1979b, 1980), light and electron microscopy (Brlansky, 1987; Brlansky et al. 1984; Garnsey et al. 1980a), and in situ immunofluorescence








3

(Brlansky et al. 1984; Tsuchizaki et al. 1978). Each of these methods has different advantages, disadvantages, and sensitivity levels. Therefore, a certain method may be used for a particular purpose. Some methods, such as ELISA, are dependable and widely used for indexing purposes (Garnsey et al. 1981a).

Recently a monoclonal antibody was developed against a decline-inducing isolate from Florida which, in ELISA, reacted specifically with several severe CTV isolates from diverse geographical areas, but not with mild isolates from the same areas (Permar et al. 1990).

Control of CTV is difficult. In those few areas of the world where CTV still is not present, quarantine and virusfree certification programs are maintained to prevent the introduction of infected budwood sources (Bar-Joseph et al. 1983, 1989). Likewise, in those areas with low CTV incidence, large scale surveys and suppression measures are carried out to reduce disease spread to other trees and locations and prolong the use of sour orange as a rootstock (Bar-Joseph et al. 1989). Once the disease becomes endemic, two situations can result: a) CTV-induced decline develops and kills plants grafted onto sour orange rootstock, whereas tolerant rootstocks do not decline, and/or b) CTV-stem pitting can affect sweet orange and/or grapefruit scions regardless of the rootstock, resulting in a loss of plant vigor and yield. Mild strain cross protection is the only known control measure








4

which is effective against stem pitting (Bar-Joseph et al. 1989; Garnsey and Lee, 1988; Lee et al. 1987a). Genetic resistance to CTV is not available in commercially acceptable scions (Bar-Joseph et al. 1989; Garnsey and Lee, 1988). The application of genetically engineered cross protection, currently effective in several other crops (Beachy et al. 1987), is an attractive possibility for CTV control in the future.

CTV has been widespread in Florida for many years and induced decline has occurred in localized areas (Garnsey and Jackson, 1975; Norman et al. 1961). However, until recently, it had not caused major losses because most of the citrus acreage had been propagated on CTV-tolerant rootstocks and because of the prevalence of mild CTV isolates which did not seriously affect trees grafted on sour orange rootstock (Brlansky et al. 1986; Garnsey et al. 1980b; Lee et al. 1987a). In the last decade the situation in Florida has changed radically. Sour orange continued to be a very popular rootstock because of cold tolerance, high fruit quality of the scion, and its tolerance to citrus blight. The high demand for plants on sour orange, plus discovery of citrus bacterial leaf spot in some nurseries (Brlansky, 1988; Garnsey, personal communication), caused nurserymen to use budwood from source trees that had not been propagated previously on sour orange. Many such trees apparently were harboring severe CTV isolates (Brlansky et al. 1986; Lee et al. 1987a). Severe dwarfing of








5

young trees propagated on sour orange has appeared in many parts of Florida. Large scale outbreaks of induced decline also have appeared in southern Florida, an area previously not affected by CTV. Losses have exceeded 50% in some plantings (Brlansky et al. 1986).

Management of CTV-induced decline in Florida is difficult. The effective use of CTV tolerant rootstocks, such as rough lemon (Citrus jambhiri Lush.), Troyer citrange {Poncirus trifoliata (L.) Raf. x C. sinensis (L.) Osb.}, Cleopatra mandarin (_. reshni Hort. ex Tanaka), sweet orange (C. sinensis) and others (Grant et al. 1961; Wallace, 1978) is diminished by their susceptibility to other diseases, most importantly citrus blight, an endemic disease of unknown etiology which is removing more than 500,000 trees from production annually (Lee et al. 1988a).

The objectives of this research were: i) To evaluate some CTV mild isolates under greenhouse conditions for cross protecting ability against the decline syndrome, and ii) To develop methods for detection of the severe CTV challenge isolate in mixed infections.














CHAPTER 2
EVALUATION OF THE PROTECTING EFFECTS OF SOME MILD FLORIDA ISOLATES OF CITRUS TRISTEZA VIRUS AGAINST THE DEVELOPMENT OF THE DECLINE SYNDROME


Introduction


Cross protection is a control strategy used to reduce losses due to plant viral diseases by the use of mild or attenuated strains of a virus which prevent the effect or expression of a related, and usually more severe, strain of the same virus (Fulton, 1986). Cross protection can be useful when the virus disease is endemic, causes great losses, and no host genetic resistance is available (Fulton, 1986; Gonsalves and Garnsey, 1989: Hamilton, 1985; Muller et al. 1982). Cross protection has been used commercially to control CTV stem pitting isolates on Pera sweet orange in Brazil (Costa and Muller, 1980; Muller, 1980), and is a part of South Africa's citrus cultivar improvement program to reduce CTVinduced stem pitting on grapefruit (DeLange et al. 1980; Garnsey and Lee, 1988). Relatively little work has been done to evaluate the potential of cross protection against the CTVinduced decline (CTV-ID) on sour orange, as most countries abandon sour orange as a rootstock when CTV-ID isolates become prevalent. However, experiments conducted in Australia



6








7

(Thornton et al. 1980), United States (Wallace and Drake, 1976; Yokomi et al. 1990), and Japan (Miyakawa, 1987) indicate that cross protection against CTV-ID isolates may be possible.

The classical approach for selecting potential cross protecting mild CTV isolates has been empirical, that is, by collecting a number of CTV isolates from outstanding trees in groves severely affected by the disease (Balaraman and Ramakrishnan, 1980; Muller and Costa, 1987). Those isolates are propagated on several scion-rootstock combinations in nursery plants and further evaluated over a period of several years under field conditions (Muller, 1980; Muller and Costa, 1977). This requires the handling and care of large numbers of plants, extensive field space, and considerable time; the results are evaluated over a period of 5-12 years. By this approach, only six mild CTV isolates out of 45 mild isolates originally selected in Brazil were useful for cross protection (Costa and Muller, 1980).

Another approach for the selection of mild CTV isolates for cross protection relies upon host effects for the strain segregation of field collected severe CTV isolates (Roistacher et al. 1988). The selection of potential cross protecting CTV isolates was made from either symptomless or recovered infected plants after extensive passage of the virus through a series of different citrus and non-citrus hosts in the greenhouse (Roistacher et al. 1987, 1988). This approach provided nine mild or attenuated CTV isolates with outstanding








8

protection against either seedling yellows or stem pitting CTV isolates from a total of 116 evaluated when challenged with the original isolates from which they were derived. This approach required extensive greenhouse work to follow the segregation of every single isolate, and some attenuated isolates showed a tendency to revert back to their severe forms.

Recently, the greenhouse evaluation of CTV-stem pitting isolates as protecting agents against the CTV-ID on sweet/sour orange combinations was reported (Miyakawa, 1987). However, CTV-stem pitting isolates as protecting agents offer limited possibilities of commercial use in Florida, particularly because stem pitting isolates still are not present and susceptible species or cultivars such as grapefruit are grown extensively.

Citrus tristeza virus causing decline on trees on sour orange rootstock is endemic in Florida. The effective use of CTV-tolerant rootstocks is diminished by their susceptibility to citrus blight. In addition, the increased popularity of sour orange despite tristeza, along with the natural prevalence of mild CTV isolates, provides the opportunity to evaluate cross protection as an alternative control strategy for CTV-ID. However, there are several considerations: i) The threat of recurrent freezes makes it difficult to reliably evaluate the cross protection potential of mild isolates under field conditions, especially in the Ridge area; ii) There is








9

a need for a method to more rapidly select and evaluate potential cross protecting mild isolates which is adaptable for screening large numbers of isolates; and iii) There is a need to differentiate mild and severe CTV isolates in a mixed infection in the same plant to aid in the evaluation and to better understand the mechanism by which cross protection functions.

The objectives of this research were to evaluate naturally occurring Florida mild CTV isolates for cross protecting ability and to develop a methodology to detect the presence of protecting and challenge CTV isolates in mixed infections. The effect of temperature on the cross protecting ability of CTV mild isolates also was studied.

Materials and Methods

Virus isolates and donor hosts. Five naturally occurring Florida CTV isolates collected from field grown sweet orange or grapefruit trees grafted on sour orange were used in these experiments after transmission by Aphis gossypii Glover. The T11a, T26, T30 and T55a isolates produce very mild symptoms and little or no stunting on Mexican lime seedlings {Citrus aurantifolia (Christm.) Swingle}, no seedling yellows on Eureka lemon {C. limon (L.) Burm.} or sour orange (c. aurantium L.) and are symptomless in sweet orange {C. sinensis

(L.) Osb.} and sweet/sour orange combinations (Garnsey et al. 1987; Lee, 1984; Yokomi and Garnsey, 1987; Yokomi et al. 1987). The T66a challenge isolate causes strong vein








10

clearing, stunting and stem pitting in Mexican lime seedlings and severe decline on sweet/sour orange combinations (Garnsey et al. 1987; Yokomi et al. 1987). The CTV isolates were propagated in either C. excelsa Wester, Madam Vinous sweet orange or Mexican lime plants and maintained in a greenhouse with mean minimum and maximum temperatures of 21 and 380C, respectively. Inoculum source tissue from donor hosts was evaluated by serological indexing (see below) to confirm the presence of CTV before being used as inoculum.

Inoculation of CTV isolates and receptor hosts. Oneyear-old Valencia sweet orange plants budded on either sour orange or C. macrophylla Wester rootstocks, were graft inoculated in the stem with each of the CTV mild isolates using three leaf pieces or blind buds per plant (Garnsey and Whidden, 1970; Garnsey et al. 1987). Inoculum tissue was sealed firmly into the receptor stems with plastic grafting tape. Three weeks later the grafting tape was removed and the plants were evaluated for survival of grafted tissue and reinoculated if the grafted tissue had not survived. After verifying by serological indexing that infection by mild isolates had taken place, a minimum of four inoculum pieces of T66a infected tissue were used to challenge the test plants, and they were reinoculated if at least two inoculum pieces were not alive 21 days post-challenge. Surviving inoculum tissue was left in place for the duration of the experiment. Inoculated receptor plants were grown in a








11

commercial potting mixture (Pro-mix BX) in five liter plastic containers, and fertilized with a mixture of NPK (20-10-20) every other week. Pest and disease management included the application of 0.300 g active ingredient (a.i.)/plant of aldicarb and 0.86 g a.i /L soil drench of ridomil twice a year. The experiment was conducted in a greenhouse with mean minimum and maximum temperatures of 21 and 380C, respectively. At least 15 plants were inoculated for each CTV isolate/scion/ rootstock combination.

A second set of one-year old Valencia sweet orange plants budded on sour orange and C. macrophylla rootstocks were inoculated with each of the CTV mild isolates and challenged with the T66a severe isolate as previously described. These plants were placed in a greenhouse with controlled mean minimum and maximum temperatures of 21 and 330C, respectively, to evaluate the effect of temperature on the cross protecting ability of the CTV isolates. At least 10 plants were inoculated per CTV isolate/scion/rootstock combination. Fertilization and plant pest and disease management was as above.

Serological tests. CTV infection and relative antigen titer of inoculated plants were determined throughout the study by the double antibody sandwich enzyme-linked immunosorbent assay (DAS-ELISA) (Bar-Joseph et al. 1979b, 1980), using polyclonal antiserum No. 1053 prepared against whole, unfixed CTV isolate T26 (R.F. Lee, unpublished). The








12

severe CTV isolate T66a was detected in the challenged plants by DAS-indirect ELISA using the MCA-13 strain specific monoclonal antibody which reacts strongly against most severe CTV isolates (Permar et al. 1990). For efficiency the serological tests were performed on both experiments at the same time.

Routinely, 0.5 g of bark, petioles and midribs of new, fully expanded tissue were finely chopped with a razor blade and ground, using a Tekmar Tissumizer, in 5 ml of phosphate buffered saline (PBS)-Tween + polyvinyl pyrrolidone {PBS = 8 mM Na2HPO4, 14 mM KH2PO4, 15 mM NaCl, pH 7.4, (+ 0.1 % Tween 20 + 2% polyvinyl pyrrolidone (PVP-40 Sigma)}. Unless stated otherwise, 200 microliters samples were used per well of the microtiter plates and three washings with PBS-Tween (phosphate buffered saline + 0.1 % Tween 20) were performed between steps. The immunoglobulins (IgG) present in the whole CTV antiserum were purified by the Protein A-Sepharose affinity method (Miller & Stone, 1978). A portion of purified immunoglobulins were conjugated to alkaline phosphatase by the glutaraldehyde method (Clark et al. 1986). Polystyrene Immulon II microtiter plates (Dynatech Laboratories) were coated with 2.0 Ag/ml of purified IgG in carbonate buffer (0.015 M NaHCO3, 0.03 M NaCO3, pH 9.6) and incubated for 6 hr at 370C. Antigen samples were added to the wells and incubated for 18 hr at 50C. Enzyme conjugate was used at a dilution of 1:1,000 in conjugate buffer (PBS-Tween + 2%








13

polyvinyl pyrrolidone + 0.2% bovine serum albumin) and incubated for 6 hr at 370C. The reaction with one mg/ml of pnitrophenyl phosphate (Sigma) in 10% triethanolamine, pH 9.8, was measured after 120 min at 405 nm (OD405) with a Labinstruments model EAR 400 AT ELISA plate spectrophotometer. Samples were considered positive when OD4, values were higher than 0.100 or three times the mean of healthy controls, whichever was greater.

For DAS-Indirect ELISA, the microtiter plates were first coated with IgG from antiserum No. 1053. Antigen samples were added as described for DAS-ELISA. The MCA-13 strain specific monoclonal antibody (hereafter MCA-13), as ascites fluid, was added at a dilution of 1:5,000 (v/v) in conjugate buffer and incubated 4 hr at 370C. After washing, goat anti-mouse IgG labeled with alkaline phosphatase (Promega) at a dilution of 1:7,500 (v/v) in conjugate buffer was added and incubated for 2 hr at 370C. The enzyme reaction was carried out as for DASELISA.

For all serological tests, two replications were used per sample. Positive controls included four mild isolates (Tlla, T26, T30, and T55a) and one severe (T66a) CTV isolate. Negative controls included extraction buffer, and similar buffer extracts from healthy C. excelsa and Valencia sweet orange plants. A standard curve prepared with purified CTV T26 isolate diluted to OD26, values of 0.04, 0.02, 0.01, 0.005, 0.0025, 0.0012, 0.0006, and 0.0003 diluted in buffer extract








14

of healthy C. excelsa was used to estimate the relative antigen concentration of test samples.

Detrimental effects of mild isolates and evaluation of cross protection. To evaluate the effect of each CTV isolate on the inoculated plants and the protecting ability of the mild isolates against the T66a challenge isolate, evaluations were made at five and ten months after the challenge inoculation with the T66a isolate. Phloem necrosis was evaluated by cutting a bark flap at the bud union and the plant tissue was examined with a hand lens for browning. A decline index was assigned for each plant. The parameters scored were stem diameter, plant growth, and foliage symptoms for decline. Each parameter was visually rated from 0 (minimum) to 3 (maximum), for a maximum cumulative score of 9 for each plant. A high decline index sometimes was accompanied by plant death. The decline index for each treatment was the average of the cumulative scores for all plants in that treatment.

Results

Antigen titers of mild and severe CTV isolates with polyclonal and MCA-13 monoclonal antibodies. The antigen titers expressed as optical density (OD405) values for the different temperature and host treatments, measured by DASELISA with polyclonal antibodies (PCA) and DAS-indirect ELISA with MCA-13, are summarized in Tables 2.1, 2.2, 2.3, and 2.4. At warm temperatures, Valencia/sour orange plants inoculated









15

with mild isolates but unchallenged with T66a gave OD405 values between 0.091 and 0.145 when analyzed with PCA. The corresponding uninoculated healthy control plants averaged 0.039. The same treatments, including the healthy controls gave values in the range of 0.011-0.025 when analyzed by DASindirect ELISA with MCA-13. Treatments pre-inoculated with mild isolates and further challenged with T66a gave OD405 values in the range of 0.130-0.217 with PCA and 0.145-0.189 with MCA-13. The control plants uninoculated with mild isolates but challenged with T66a gave values of 0.174 with PCA and 0.214 with MCA-13 (Table 2.1).

Also at warm temperatures, the Valencia/macrophylla plants inoculated with the mild isolates and unchallenged with T66a, gave OD405 values between 0.092 and 0.164 with PCA. The value for the corresponding uninoculated healthy control plants was 0.040. The same treatments, including the healthy controls gave values in the range of 0.018-0.037 when analyzed with MCA-13. Treatments pre-inoculated with mild isolates and challenged with the T66a gave values in the range of 0.1850.311 with PCA and 0.104-0.305 with MCA-13. The control plants uninoculated with mild isolates but challenged with T66a gave values of 0.194 with PCA and 0.251 with MCA-13 (Table 2.2). The T30 isolate was not evaluated in this portion of the experiment because not enough plants were available. The plants pre-inoculated with the T11a isolate were not protected and declined and died before the








16



Table 2.1 Relative antigen titer of citrus tristeza virus (CTV) mild isolates in plants unchallenged and challenged by the T66a severe isolate: I. Warm temperature, Valencia/sour orange.



Unchallenged1 Challenged/2/

CTV OD 3/ OD isolate Polyclonal MCA-13 Polyclonal MCA-13

Tlla 0.101ab/ 0.024 a 0.217 a 0.175 a

T26 0.123 ab 0.025 a 0.130 a 0.145 a

T30 0.091 ab 0.018 ab 0.212 a 0.189 a

T55a 0.145 a 0.014 b 0.194 a 0.181 a

Healthy 0.039 b 0.011 b 0.174 a 0.214 a or control
plants uninoculated
with mild isolate



1' One-year-old plants were graft inoculated with leaf
pieces under bark flaps on the stem from donor plants
infected with the indicated mild CTV isolates.
After verifying virus infection with mild isolates by DAS-ELISA with polyclonal antibodies, the challenged plants were graft inoculated similarly with the T66a
severe isolate.

3J Optical density at 405 nm (OD4.) was measured after 120 min
of reaction.

Serological detection was carried out by DAS-ELISA with
polyclonal antisera and DAS-indirect ELISA with MCA-13
severe strain specific monoclonal antibody six months after inoculation. Value is the average of duplicate
assays of the corresponding plants in Table 2.5.

Numbers in the same column followed by different letters
are statistically different by Duncan's test (P A 0.05).








17

Table 2.2 Relative antigen titer of citrus tristeza virus (CTV) mild isolates in plants unchallenged and challenged by the T66a severe isolate: II. Warm temperature, Valencia/macrophylla.


Unchallenged1/ Challenged1/21

CTV OD3/ ODO isolate Polyclonal MCA-13 Polyclonal MCA-13

Tlla 0.092yal/ 0.026 a -

T26 0.132 a 0.030 a 0.185 a 0.104 a

T30 NE7 NE NE NE

T55a 0.164 a 0.037 a 0.311 a 0.305 a

Healthy 0.040 a 0.018 a 0.194 a 0.251 a or control
plants uninoculated
with mild isolate



' One-year-old plants were graft inoculated with leaf
pieces under bark flaps on the stem from donor plants
infected with the indicated mild CTV isolates.
' After verifying virus infection with mild isolates by
DAS-ELISA with polyclonal antibodies, the challenged plants were graft inoculated similarly with the T66a
severe isolate.

Optical density at 405 nm (OD405) was measured after 120 min
of reaction.

/ Serological detection was carried out by DAS-ELISA with
polyclonal antisera and DAS-indirect ELISA with MCA-13
severe strain specific monoclonal antibody six months after inoculation. Value is the average of duplicate
assays of the corresponding plants in Table 2.6.

Numbers in the same column followed by different letters
are statistically different by Duncan's test (P < 0.05).
- = severely diseased plants died before serological
evaluation.

NE = treatment not evaluated.








18

serological evaluation was made five months after inoculation (Table 2.2).

The OD405 values were generally higher for the test plants grown at cooler temperatures. The Valencia/sour orange plants inoculated with mild isolates and unchallenged with T66a gave OD405 values between 0.138 and 0.405 when analyzed with PCA. The corresponding healthy plants averaged 0.039. The same treatments, including healthy controls, gave values in the range of 0.016-0.078 with MCA-13. Treatments pre-inoculated with mild isolates and challenged with T66a gave OD405 values in the range of 0.174-0.548 with PCA and 0.047-0.477 the MCA13. The control plants uninoculated with mild isolates but challenged with T66a gave OD405 values of 0.292 with PCA and

0.283 with MCA-13 (Table 2.3).

Also at cooler temperatures the Valencia/macrophylla plants inoculated with mild isolates and unchallenged with T66a, gave OD405 values between 0.335 and 0.361 with PCA. The value for the corresponding uninoculated healthy control plants was 0.036. The same treatments, including healthy controls, gave values in the range of 0.013-0.075 when analyzed with MCA-13. Treatments pre-inoculated with mild isolates and challenged with T66a gave values in the range of 0.406-0.456 with PCA and 0.166-0.320 with MCA-13. The control plants uninoculated with mild isolates but challenged with T66a gave values of 0.432 with PCA and 0.421 with MCA-13 (Table 2.4). The Tlla and T30 isolates were not evaluated for








19



Table 2.3 Relative antigen titer of citrus tristeza virus (CTV) mild isolates in plants unchallenged and challenged by the T66a severe isolate: III. Cool temperature, Valencia/sour orange.


UnchallengedI/ Challenged1/2/

CTV 01) 3/ OD4 isolate Polyclonal MCA-13 Polyclonal MCA-13

Tlla 0.404k' P 0.060 a 0.548 a 0.477 a

T26 0.405 a 0.078 a 0.195 b 0.055 c

T30 0.138 a 0.054 a 0.174 b 0.047 c

T55a 0.308 a 0.055 a 0.276 b 0.057 c

Healthy 0.039 b 0.016 a 0.292 b 0.283 b or control
plants uninoculated
with mild isolate



/ One-year-old plants were graft inoculated with leaf
pieces under bark flaps on the stem from donor plants
infected with the indicated mild CTV isolates.
2 After verifying virus infection with mild isolates by
DAS-ELISA with polyclonal antibodies, the challenged plants were graft inoculated similarly with the T66a
severe isolate.

1 Optical density at 405 nm (0D405) was measured after 120 min
of reaction.

y Serological detection was carried out by DAS-ELISA with
polyclonal antisera and DAS-indirect ELISA with MCA-13
severe strain specific monoclonal antibody six months after inoculation. Value is the average of duplicate
assays of the corresponding plants in Table 2.7.

st Numbers in the same column followed by different letters
are statistically different by Duncan's test (P 5 0.05).








20



Table 2.4 Relative antigen titer of citrus tristeza virus (CTV) mild isolates in plants unchallenged and challenged by the T66a severe isolate: IV. Cool temperature, Valencia/macrophylla.



Unchallenged"/ ChallenQed1/2/

CTV OD3/ OD isolate Polyclonal MCA-13 Polyclonal MCA-13

Tlla NE NE NE NE

T26 0.361 a 0.049 a 0.406 a 0.166 a

T30 NE NE NE NE

T55a 0.335 a 0.075 a 0.456 a 0.320 a

Healthy 0.036 b 0.013 b 0.432 a 0.421 a or control
plants uninoculated
with mild isolate



' One-year-old plants were graft inoculated with leaf
pieces under bark flaps on the stem from donor plants
infected with the indicated mild CTV isolates.
' After verifying virus infection with mild isolates by
DAS-ELISA with polyclonal antibodies, the challenged plants were graft inoculated similarly with the T66a
severe isolate.

' Optical density at 405 nm (OD405) was measured after 120 min
of reaction.

NE = treatment not evaluated.

' Serological detection was carried out by DAS-ELISA with
polyclonal antisera and DAS-indirect ELISA with MCA-13
severe strain specific monoclonal antibody six months after inoculation. Value is the average of duplicate
assays of the corresponding plants in Table 2.8.

W Numbers in the same column followed by different letters
are statistically different by Duncan's test (P ! 0.05).








21

the Valencia/macrophylla combination (Table 2.4) because not enough plants were available.

From the standard curve prepared with purified T26 (Fig. 2.1), it was estimated that an OD405 value of 0.638 was approximately equivalent to 20 Ag/ml of CTV antigen, assuming an extinction coefficient of 2.0 (Gonsalves et al. 1978). Therefore, there was an average of 3.1 Ag of CTV antigen per every 100 mg of tissue for each 0.100 OD405 value in the test samples.

Detrimental effects of mild isolates and evaluation of cross protection. The protecting effect of mild isolates was evaluated on the basis of their ability to prevent the detrimental effects on stem diameter, plant growth and foliage symptoms caused under greenhouse conditions by the T66a severe decline isolate. The detrimental effects of mild isolates alone also were evaluated. The number of plants and the scores for the decline index established in every treatment are shown in Tables 2.5, 2.6, 2.7, and 2.8. At warm temperatures, Valencia/sour orange plants inoculated with mild isolates and unchallenged with T66a, and the healthy uninoculated controls, gave overall decline index values in the range of 0.0 and 1.5. Whereas, the plants pre-inoculated with mild isolates and challenged with T66a showed higher decline index values of 6.3, 2.6, 4.2, and 5.5 for the T11a, T26, T30, and T55a mild isolates, respectively. The control plants uninoculated with mild isolates but challenged with









22








0.7 T m
-20 *
S Tf3*
0.6

0 " t 0.5 15 1

0.4
0
to 10 C C 0.3

- 0.2


0
0 0 0 0.01 0.02 0.03 0.04 0.05
Purified virus (Optical Density 260 nm)



Figure 2.1 Plot of purified citrus tristeza virus (CTV) against optical density. Bark of healthy Citrus excelsa (0.5 g) tissue was ground in 5.0 ml of phosphate buffered saline, pH 7.6, + 0.05% Tween + 2% polyvinyl pyrrolidone and mixed with purified CTV T26 isolate to give the desired optical density at 260 nm (OD260). DAS-ELISA was performed as described in materials and methods. An extinction coefficient of 2.0 was assumed (Gonsalves et al. 1978) to estimate the relative virus concentration.








Table 2.5 Effect of citrus tristeza virus (CTV) mild isolates on the development of the CTV decline syndrome in plants unchallenged and challenged by the T66a severe isolate: I. Warm temperature, Valencia/sour orange.


Unchallenged/ Challengedl/2/

CTV No. of Decliq3 No. of Decline No. of dead isolate plants index-" plants index plants 10 months after challenge

Tlla 4 0.0 8 6.3 3/8

T26 4 1.5 5 2.6 0/5

T30 5 1.0 4 4.2 1/4

T55a 5 0.0 7 5.5 2/7

Healthy 5 1.4 6 4.0 1/6 or control
plants uninoculated
with mild isolate


One year old plants were graft inoculated with leaf pieces under bark flaps on the
stem from donor plants infected with the indicated mild CTV isolates.

2/ After verifying virus infection with mild isolates by DAS-ELISA with polyclonal
antibodies, the challenged plants were graft inoculated similarly with a T66a severe
isolate.

3/ Decline index is the average per treatment of the cumulative score given by visual
readings on three criteria: stem diameter, growth reduction, and foliage decline
symptoms, where 0 = healthy vigorous plant to 3 = severely diseased. Minimum index
= 0, maximum index = 9. A high decline index sometimes was accompanied by plant
death.








24

T66a averaged a decline index of 4.0 (Table 2.5). The lowest number of dead plants in the experiment (0/5) was obtained when T26 was the protecting isolate (Table 2.5). At warm temperatures, Valencia/macrophylla plants inoculated with mild isolates, including the healthy uninoculated controls gave decline indexes in the range of 1.7-2.2. In comparison, the plants pre-inoculated with mild isolates and challenged with T66a, showed higher decline index values in the range of 2.29. The decline index scores for the control plants uninoculated with mild isolates but challenged with T66a averaged 6.8 (Table 2.6). The lowest number of dead plants occurred when the T26 (0/5) and T55a (0/2) were the protecting isolates (Table 2.6).

At cool temperatures, the decline index values for Valencia/sour orange plants inoculated with mild isolates were in the range of 1.0-3.2. The index of 3.0 for the healthy uninoculated controls indicated the generally reduced growth rate of plants at cool temperatures. The decline index values for plants pre-inoculated with mild isolates and challenged with T66a ranged from 5.0 to 7.5. The corresponding control plants uninoculated with mild isolates but challenged with T66a averaged a decline index of 7.5 (Table 2.7). The lowest number of dead plants occurred when T26 (0/2) and T55a (1/4) were the protecting isolates (Table 2.7).

Also at cool temperatures, the Valencia/macrophylla plants inoculated with T26 and T55a isolates and the





Table 2.6 Effect of citrus tristeza virus (CTV) mild isolates on the development of the CTV decline syndrome in plants unchallenged and challenged by the T66a severe isolate: II. Warm temperature, Valencia/macrophylla.

UnchallenQedl/ Challengedl/2/

CTV No. of Decliqq No. of Decline No. of dead isolate plants index-" plants index plants 10 months after challenge
Tlla 1 2.0 3 9 3/3

T26 4 1.7 5 4.2 0/5

T30 NE4/ NE NE NE NE

T55a 2 2.0 2 2.2 0/2

Healthy 5 2.2 6 6.8 3/6 or control
plants uninoculated
with mild isolate

i/ One year old plants were graft inoculated with leaf pieces under bark flaps on the
stem from donor plants infected with the indicated mild CTV isolates.

2/ After verifying virus infection with mild isolates by DAS-ELISA with polyclonal
antibodies, the challenged plants were graft inoculated similarly with a T66a
severe isolate.

3/ Decline index is the average per treatment of the cumulative score given by visual
readings on three criteria: stem diameter, growth reduction, and foliage decline
symptoms, where 0 = healthy vigorous plant to 3 = severely diseased. Minimum index
= 0, maximum index = 9. A high decline index sometimes was accompanied by plant
death.

4/ NE = treatment not evaluated.







Table 2.7 Effect of citrus tristeza virus (CTV) mild isolates on the development of the CTV decline syndrome in plants unchallenged and challenged by the T66a severe isolate: III. Cool temperature, Valencia/sour orange.

UnchallenQedl/ Challengedl/2/
CTV No. of Decliq No. of Decline No. of dead isolate plants index- plants index plants 10 months after challenge
Tlla 3.2 4 7.5 3/4

T26 1 1.0 2 5.5 0/2

T30 2 3.0 3 6.6 2/3

T55a 4 2.7 4 5.0 1/4

Healthy 1 3.0 2 7.5 1/2 or control
plants uninoculated
with mild isolate

One year old plants were graft inoculated with leaf pieces under bark flaps on the
stem from donor plants infected with the indicated mild CTV isolates.

2/ After verifying virus infection with mild isolates by DAS-ELISA with polyclonal
antibodies, the challenged plants were graft inoculated similarly with a T66a
severe isolate.
3/ Decline index is the average per treatment of the cumulative score given by visual
readings on three criteria: stem diameter, growth reduction, and foliage decline
symptoms, where 0 = healthy vigorous plant to 3 = severely diseased. Minimum index
= 0, maximum index = 9. A high decline index sometimes was accompanied by plant
death.








27

uninoculated healthy controls showed decline index scores of 5.0, 4.3, and 3.0, respectively. The plants pre-inoculated with the T26 and T55a isolates and challenged with T66a showed scores of 4.2 and 3.5, respectively. The uninoculated controls challenged with T66a showed a decline index of 5.4 (Table 2.8). No dead plants were scored in this experiment 10 months after the challenge inoculations (Table 2.8).

Phloem necrosis at the bud union was not observed in any treatment at either temperature at five or ten months after the challenge inoculation (not shown).

Discussion

In this study four different naturally occurring Florida mild CTV isolates were evaluated for their cross-protecting ability against the development of the CTV-induced decline in two susceptible scion/rootstock combinations. DAS-ELISA with polyclonal antisera was used to determine the total antigen titer in plants inoculated with mild isolates and those also challenged with the severe T66a isolate. The MCA-13 monoclonal antibody was evaluated in DAS-indirect ELISA for quantitation of the T66a severe challenge isolate in mixed infections.

There were some limitations in the transmissibility of CTV by leaf piece grafts from the different hosts used to propagate the CTV isolates. At the beginning of the work most of the CTV isolates had been propagated in Citrus excelsa plants, which has been reported as an excellent propagation






Table 2.8 Effect of citrus tristeza virus (CTV) mild isolates on the development of the CTV decline syndrome in plants unchallenged and challenged by the T66a severe isolate: IV. Cool temperature, Valencia/macrophylla.


UnchallenQedl/ Challenged1/2/

CTV No. of Declinq No. of Decline No. of dead isolate plants index-" plants index plants 10 months after challenge

Tlla NE4/ NE NE NE NE

T26 4 5.0 5 4.2 0/5

T30 NE NE NE NE NE

T55a 3 4.3 4 3.5 0/4

Healthy 3 3.0 5 5.4 0/5 0 or control
plants uninoculated
with mild isolate


One year old plants were graft inoculated with leaf pieces under bark flaps on the
stem from donor plants infected with the indicated mild CTV isolates.

2/ After verifying virus infection with mild isolates by DAS-ELISA with polyclonal
antibodies, the challenged plants were graft inoculated similarly with a T66a
severe isolate.

3/ Decline index is the average per treatment of the cumulative score given by visual
readings on three criteria: stem diameter, growth reduction, and foliage decline
symptom, where 0 = healthy vigorous plant to 3 = severely diseased. Minimum index
= 0, maximum index = 9. A high decline index sometimes was accompanied by plant
death.

4/ NE = treatment not evaluated.








29

host for purification purposes of CTV (Lee, et al. 1987b, 1988b). However, it had never been used for leaf graft transmission experiments. At least three inoculations with C. excelsa tissue infected with mild isolates were made unsuccessfully, even though the inoculated tissue survived for at least four weeks post-inoculation and was left in place for several months. It was necessary to switch to Madam Vinous sweet orange and/or Mexican lime plants infected with the mild isolates as inoculum sources before a minimum of 3 or 4 plants per treatment were positively infected with mild isolates as determined by DAS-ELISA. Furthermore, some treatments, mostly Valencia/macrophylla, were not evaluated because of the lack of replications because not enough plants were available. This originated the need to design another separate experiment, described in Chapter 3, to determine the effectiveness of different citrus species as donor hosts for graft transmission of the virus.

At warm temperatures, plants inoculated with mild isolates but unchallenged with T66a had relatively low OD405 values in the range of 0.095-0.145 (Tables 2.1 and 2.2). At cool temperatures, with few exceptions, were commonly higher in the range of 0.300-0.400 (Tables 2.3 and 2.4). At warm temperatures there were some treatments that gave OD405 values lower than 0.100, which may be interpreted as negative reactions. However, those values were the averages of the OD405 readings of every treatment. Thus a single plant with a








30

very low OD405 value could cause the whole treatment average to be lower than 0.100. It has been suggested that a high titer in a plant infected with a mild CTV isolate may be a relative estimate of the protecting ability of mild isolates in crossprotection experiments (Koizumi and Kuhara 1984; Lee et al. 1987a).

Some differences were found in the reaction of the MCA13 monoclonal antibody in the different treatments and temperatures evaluated. At warm temperatures plants preinoculated with mild isolates but unchallenged, gave low OD405 values in the range of the uninoculated control plants (Table 2.1 and 2.2). Likewise, at cool temperatures, the OD405 values obtained with the MCA-13 with mild isolates were slightly higher than those obtained at warm temperatures (Tables 2.3 and 2.4). This could be interpreted that the MCA-13 monoclonal antibody may react to some extent with mild isolates when they are above a certain titer in the plants. However, the OD405 values were always lower than 0.100, which was considered a negative reaction.

When plants pre-inoculated with mild isolates and further challenged with the T66a severe isolate were analyzed with the MCA-13 in DAS-indirect ELISA (Tables 2.1 and 2.2), the OD405 values were generally lower than the unprotected challenged control plants, even though the differences usually were not statistically significant (with the exception T55a in Table 2.2). At cool temperatures, a similar phenomenon was observed








31

(Tables 2.3 and 2.4). In this situation, the T26 and T30 isolates generally gave the lowest OD405 values of all treatments evaluated. This suggests that the mild isolates, especially T26 and T30, but also T55a to some degree, prevented or reduced the multiplication of the T66a challenge isolate. Thus these mild isolates were apparently working as cross-protecting agents.

These results provide further evidence of the usefulness of the MCA-13 monoclonal antibody to detect the presence of severe CTV isolates in mixed infections. The MCA-13 has been previously used in other studies to evaluate the presence of severe isolates in field cross-protection experiments (RochaPefia et al. 1990; Yokomi et al. 1990). Determination of the OD405 readings when the MCA-13 monoclonal antibody is used to detect the presence of the severe isolate in cross-protection experiments provides a measurable parameter to estimate the ability of mild isolates to prevent the establishment of severe challenge isolates in such experiments.

Some differences were found in the effects of CTV mild isolates on the inoculated plants and in their ability to prevent detrimental effects caused by the T66a challenge isolate at different temperatures. At warm temperatures, the effect of the CTV mild isolates on performance of both Valencia/sour orange and Valencia/macrophylla were negligible. The decline indexes for the healthy uninoculated control plants were nearly equal to or greater than those inoculated








32

only with mild isolates (Table 2.5 and 2.6). In regard to Valencia/sour orange plants pre-inoculated with mild isolates and challenged with the T66a severe isolate, the lowest decline index (2.6) and lowest number of dead plants (0/5) was obtained with the T26 isolate. Whereas, the highest decline index (6.3) and highest number of dead plants (3/8) was obtained with the Tlla isolate. A decline index of 4.0 and 1/6 dead plants were scored with the uninoculated and challenged control plants (Table 2.5). Also at warm temperatures, Valencia/macrophylla plants pre-inoculated with mild isolates and challenged with the T66a severe isolate, the T26 and T55a isolates obtained the lowest decline index (4.2 and 2.2) and lowest number of dead plants (0/5 and 0/2), respectively.

At cooler temperatures, there were some differences in the effect of the CTV mild isolates on growth of both Valencia/sour orange and Valencia/macrophylla. The decline indexes for the healthy uninoculated control plants were variable and ranged from values below to above those obtained with plants inoculated only with mild isolates (Table 2.7 and 2.8). In general, there was a remarkable growth reduction effect at cool temperatures even on healthy uninoculated control plants. In this regard Valencia/sour orange plants pre-inoculated with mild isolates and challenged with the T66a severe isolate, the T26 and T55a isolates obtained the lowest decline index (5.5 and 5.0) and lowest number of dead plants








33

(0/2 and 1/4), respectively. Whereas, the highest decline index (7.5) and highest number of dead plants (3/4) were obtained with the T11a isolate. A decline index of 7.5 and 1/2 dead plants were scored with the uninoculated and challenged control plants (Table 2.7). At cool temperatures, Valencia/macrophylla plants pre-inoculated with mild isolates and challenged with the T66a severe isolate, had decline index scores similar to the plants inoculated only with mild isolates. No dead plants were obtained in this portion of the experiment (Table 2.8).

Of all treatments evaluated at both temperatures, the T26 isolate, and also T55a to some degree, obtained the lowest decline index scores and lowest number of dead plants as compared with the uninoculated challenged control plants. This provided further evidence of their cross-protecting effect, especially the T26 isolate against the development of the CTV-ID syndrome.

Of special interest was the high decline index scores and number of dead plants in plants pre-inoculated with the T11a mild isolate and further challenged with the T66a isolate. It seemed that the combination of T11a and the T66a isolates produced a more severe reaction on the challenged plants than that caused by the T66a isolate alone in the unprotected control plants. The lack of cross-protecting ability of the T11a isolate has been previously reported (Yokomi et al. 1987).








34

The relatively low decline index scores and low occurrence of dead plants at both temperatures in the unprotected control plants challenged with the T66a isolate, indicates that under greenhouse conditions, the use of one single challenge isolate might not be sufficient to obtain an appropriate rate of decline in a short time basis. It is well documented that CTV occurs naturally as mixtures of isolates or strains with diverse biological properties (Garnsey et al. 1987, McClean, 1974). The T66a severe isolate was originally isolated from an infected field source, and subsequently aphid transmitted to avoid contamination with other viruses (Garnsey et al. 1987; Yokomi and Garnsey, 1987). It is a possibility that part of the original decline components from the field source could have been lost in the subsequent aphid transmissions. To overcome this possibility, it may be advisable in the future to use a mixture of several severe isolates as a challenge to enhance the possibility of obtaining an appropriate occurrence of decline under greenhouse conditions. Another alternative could be the use of higher populations of aphids (50 or 100) to obtain a more complete complex of CTV severe isolates from field samples.

Cross-protection using mild virus isolates as a strategy to reduce losses due to CTV has been used in Brazil (Costa and Muller, 1980; Muller, 1980), South Africa (DeLange et al. 1980; Garnsey and Lee, 1988), Japan (Ieki, 1989; Koizumi, 1986), India (Balaraman and Ramakrishnan, 1980) and Australia








35

(Cox et al. 1976; Fraser et al. 1968), against stem-pitting isolates either in orange, grapefruit, and/or acid lime. Several approaches have been reported for the evaluation of mild isolates under greenhouse conditions. However, these approaches have been addressed mostly to the evaluation of the cross-protecting effect of mild isolates against stem pitting and have included only the host reaction of Mexican lime, sweet orange, or grapefruit seedlings (Roistacher et al. 1987, 1988; Van Vuuren and Noll, 1987). Another approach where the challenge inoculations are made by using insect vectors to screen mild isolates (Yokomi et al. 1987) has not been extensively used.

The results of this work provide further evidence that a) the cross-protection against the CTV-induced decline on sweet/sour orange combinations may be possible; b) the preliminary evaluation of mild isolates under greenhouse conditions can be made in a relatively short time basis, and c) the severe challenge isolate can be detected by using the MCA-13 strain specific monoclonal antibodies. The recent report of Miyakawa (1987) about the feasibility of crossprotection on sweet/sour orange combinations, supports these conclusions.

Considering the relatively high virus titer found at cool temperatures, it is advisable to propagate the donor plants at temperatures in the range of 21-330C to better guarantee a high percentage of CTV transmission to the receptor plants.








36

Likewise, a mixture of several severe isolates should be used as the challenge virus source to give a better evaluation of

decline symptoms under greenhouse conditions.

The methodology described herein offers the following

advantages: i) Depending upon space availability, large numbers of mild isolates can be evaluated uniformly in a time period of 18 to 24 months: from six to twelve months to get the one-year-old plants infected with the mild isolates and verification of infection by serology, two months for challenge and ten months for final evaluation; ii) The availability of the MCA-13 monoclonal antibody provides a useful tool to detect the presence of a severe isolate in the challenged plants, and at the same time allows an estimate of the relative ability of mild isolates to prevent the establishment of the severe isolate in the challenged plants; iii) Mild isolates can be evaluated in grenhouses without the risks that represent the threat of recurrent freezes especially in Florida in recent years, the lack of an appropriate natural challenge pressure (vector or severe isolate), and the effect of some other devastating diseases (i.e. greening or blight) that can hamper the reliable evaluation of cross-protection experiments under field conditions. Some limitations in the methodology can be also visualized. The use of leaf piece grafts is not always highly efficient to transmit CTV from the donor propagation hosts to the receptor test plants (see Chapter 3). This leaves the




!








37

possibility that a lack of transmissibility by leaf piece grafts of the CTV challenge isolate, can be interpreted erroneously as protecting effect by mild isolates. On the other hand, the inoculum tissue with the severe isolate is left in place to enhance the probability of graft-transmission in the challenged plants. This would supply a permanent source of the severe isolate against the mild isolates which may provide a stronger challenge pressure than happens under natural conditions. If this occurs, mild isolates with potential protecting ability under natural challenge conditions could be underestimated or overlooked.














CHAPTER 3
EFFECTIVENESS OF CITRUS SPECIES AS DONOR HOSTS FOR
GRAFT TRANSMISSION OF CITRUS TRISTEZA VIRUS


Introduction


Citrus tristeza virus (CTV) has long been known to be transmitted by budding and by different grafting procedures (Bennett and Costa, 1949; Bar-Joseph et al. 1979a; Bar-Joseph and Lee, 1990). In 1951, Wallace experimentally transmitted CTV by placing small portions of donor leaf or bark tissue under a flap of bark on receptor plants. By this method, CTV was transmitted from many field sources of sweet orange to Mexican lime receptor plants, and from Mexican lime to healthy sweet orange plants (Wallace, 1951). Schwartz (1968) transmitted CTV by connecting the distal portion of the leaf of infected plants to a matching proximal part on a leaf of a receptor plant. By this method, CTV was transmitted to 9 of 20 Mexican lime plants, but transmission was obtained only when callus formation occurred between grafted tissues; also, older and dark green leaves were a better source than younger leaves for both callus formation and virus transmission.

Cohen (1972) described a method for CTV transmission in citrus by grafting triangular leaf pieces into triangular holes cut in the leaves of receptor plants. He transmitted


38








39

CTV from Meyer lemon to 25 of 27 Mexican lime and sour orange seedlings when leaf pieces contained midribs. However, the efficiency of transmission decreased more than 50% when grafts did not include the leaf midrib. A modification of Cohen's procedure, called "leaf-disc grafting" (Blue et al. 1976) involved the use of circular leaf pieces 6 mm in diameter cut from the midrib area of a donor plant leaf and placed into a corresponding hole in the receptor plant leaf. The midrib of the donor tissue is aligned with that of the receptor leaf, and grafts are held in place with transparent tape. This method was as successful as bud inoculation for transmitting many CTV isolates to Mexican lime plants from different citrus species, and was more efficient for transmitting mild CTV isolates. The leaf-disc method was used for routine indexing in the citrus budwood certification program in California (Calavan et al. 1978).

Another method involving the use of leaf piece grafts was reported by Garnsey and Whidden (1970). Rectangular leaf pieces from infected plants were inserted under corresponding rectangular bark flaps cut in the stem of receptor hosts. This procedure has been used widely for many years with CTV and other citrus viruses, and it has been used in the characterization of the biological properties of diverse worldwide collection of CTV isolates (Garnsey et al. 1987). Leaf piece grafts are especially advantageous when large








40

numbers of plants are to be inoculated with limited sources of inoculum (Garnsey and Whidden, 1970).

During several experiments with CTV in Florida (Chapter 2; Rocha-Pefla et al. 1990) large numbers of plants were inoculated by leaf piece grafts with several CTV isolates that were propagated in different citrus hosts. There were notable differences in the efficiency of transmission of some CTV isolates from different donor hosts, and in some cases no transmission was achieved even after repeated inoculations. The objectives of this research were to evaluate the effect of different citrus hosts on the efficiency of graft transmission of CTV, and to determine the relative distribution of the virus in different host tissues.

Materials and Methods

Virus isolates and donor hosts. Three isolates of CTV, T26, T30, and T66a, were used throughout the study. They have been described previously (Garnsey et al. 1987; Lee, 1984; Yokomi and Garnsey, 1987). Virus isolates were propagated in Citrus excelsa Wester, Mexican lime {C. aurantifolia (Christm.) Swingle} and Madam Vinous sweet orange {C. sinensis

(L.) Osb.} plants, herein referred to as donor hosts, maintained in a greenhouse with mean minimum and maximum temperatures of 210 and 330C, respectively. Inoculum tissue from donor hosts was evaluated by serological indexing by the double antibody sandwich enzyme-linked immunosorbent assay (DAS-ELISA) (see below) to verify the presence of CTV before








41

being used as inoculum. All donor host plants were known to be CTV infected for at least one year before the study was started.

Grafting procedures and receptor hosts. Rectangular leaf and bark pieces of about 3 X 15 mm were cut from donor hosts with a sharp knife and inserted under corresponding bark flaps cut on the stem of one-year-old Madam Vinous sweet orange, Mexican lime, and grapefruit (C. paradisi Macf.) plants, herein referred to as receptor hosts. A portion of the grafted tissue (2-3 mm) was left exposed at the top of bark flaps to monitor tissue survival at 21 days post-inoculation. A minimum of five plants of each receptor host were each inoculated with 4 pieces of either leaf or bark tissue for every donor host/virus isolate combination tested. Serological indexing by the double antibody sandwich enzymelinked immunosorbent assay (DAS-ELISA) (see below) was carried out on receptor hosts at three and five months postinoculation.

Inoculated receptor plants were grown in a commercial potting mixture (Pro-mix BX) in three liter plastic containers, and fertilized with a mixture of NPK (20-10-20) every other week, and given disease and pest management as described in Chapter 2.

Virus distribution and antiQen concentration in host tissues. Individual Madam Vinous sweet orange and C. excelsa plants infected with CTV isolates T26 or T66a were used to








42

study the relative distribution and antigen concentration of the virus in different tissues of the host plant. Bark, petioles, midribs, and leaf blades of four individual branches of each test plant, were assayed individually by DAS-ELISA. At least four replications were assayed for every host/virus isolate combination tested.

Purification of CTV. Citrus tristeza virus was purified from tender new tissue of C. excelsa greenhouse grown plants infected with the T26 isolate, by the Driselase method (Garnsey et al. 1981b; Lee et al. 1988b). The final virus preparations were adjusted with 0.05 M Tris buffer to optical density values (OD0) of 0.4 and stored in one ml aliquots at

-180C.

Serological tests. The double antibody sandwich enzymelinked immunosorbent assay (DAS-ELISA) (Bar-Joseph et al. 1979b, 1980) was conducted with polyclonal antiserum no. 1053 prepared against whole, unfixed CTV isolate T26 (R.F. Lee, unpublished). Polystyrene Immulon II microtiter plates (Dynatech Laboratories) were used. Unless otherwise stated, 200 microliters were used per well of the microtiter plates and, three washings with phosphate buffered saline (PBS)-Tween {PBS = 8 mM Na2HPO4, 14 mM KH2PO4, 15 mM NaCl, pH 7.4, (+ 0.1 % Tween 20)} were performed between steps. Host tissue (bark, petioles, midribs, etc.) was chopped finely with a razor blade and ground in a Tekmar Tissumizer in extraction buffer (PBSTween + 2% polyvinyl pyrrolidone (PVP-40 Sigma) at a 1:20








43

(w/v) dilution. Microtiter plates were coated with 2.0 Ag/ml of purified CTV specific IgG in carbonate buffer (0.015 M NaHCO3, 0.03 M NaCO3, pH 9.6) and incubated for 6 hr at 370C. Antigen samples were added to the wells and incubated for 18 hr at 50C. CTV specific IgG conjugated to alkaline phosphatase was used at a dilution of 1:1,000 in conjugate buffer (PBS-Tween + 2% PVP + 0.2% bovine serum albumin) and incubated for 4 hr at 370C (Bar-Joseph et al. 1979b, 1980). The reaction with one mg/ml of p-nitrophenyl phosphate (Sigma) in 10% triethanolamine, pH 9.8, was measured at 120 min at 405 nm (OD405) with a Bio-Tek EL-307 ELISA plate spectrophotometer. Samples were considered positive when OD405 values were higher than 0.100 or three times the mean of healthy controls, whichever was greater. There were two replications per sample in each microtiter plate. To estimate the relative CTV concentration in test samples, a standard curve prepared by diluting purified CTV T26 to OD2, values of 0.04, 0.02, 0.01, 0.005, 0.0025, 0.00125, and 0.0006 in a PBS-Tween + PVP buffered extract of bark of healthy Citrus excelsa it was included as a positive control in every test. Negative controls included PBS-Tween + 2% PVP, conjugate buffer, and extract from healthy g. excelsa, Madam Vinous sweet orange, Mexican lime and grapefruit plants.

Results

Graft transmission of citrus tristeza virus isolates.

At 21 days post-inoculation the survival rate of grafted








44

tissue in the whole experiment was 83 and 66 percent for leaf and bark pieces, respectively. Overall at least one of the four grafts survived in 92 and 90 percent of the receptor plants inoculated with leaf or bark pieces, respectively. In calculating the percent of virus transmission for each donor/receptor host/virus isolate combination, only those plants with at least one (of four) surviving inoculum piece were taken into account. Thus, overall there was a greater efficiency of transmission with leaf pieces (89.2%) than with bark pieces (75.6%) for the whole experiment (Table 3.1). There were three plants of 270 in the entire experiment, one Mexican lime and two grapefruit that became infected even though no successful graft was scored 21 days postinoculation.

The overall rate of transmission of CTV by graft inoculation for each donor/receptor host combination is shown in Table 3.2. With C. excelsa as donor host there was 72.4%, 86.9%, and 60.7% transmission to Madam Vinous, Mexican lime and grapefruit, respectively. With Mexican lime as donor host, there was 93.1%, 76.9% and 89.3% transmission to Madam Vinous, Mexican lime and grapefruit, respectively. With Madam Vinous as donor host there was 86.7%, 100%, and 84.6% transmission to Madam Vinous, Mexican lime and grapefruit respectively. The overall average of transmission was 72.5% from C. excelsa, 85.2% from Mexican lime, and 90.6% from Madam Vinous (Table 3.2). Statistical analysis showed significant











45


Table 3.1 Transmission of citrus tristeza virus by graft inoculation between selected citrus hosts: I. Efficiency of leaf and bark pieces as inoculum.


% plants with
Inoculum Inoculum at least one % transmission tissue survival% successful graft

leaf 83.0~ay 92.0 89.2 a

bark 66.0 b 90.0 75.6 b


1/ Measured at 21 days post-inoculation.

2/ Percent transmission to plants with at least one inoculum
piece (of four) alive, measured serologically by DAS-ELISA at
3 and 5 months post-inoculation. Number indicates overall
transmission for all donor/receptor/virus isolate combinations.

A total of 270 plants (135 each) were inoculated with four pieces of either leaf or bark tissue. Number indicates overall survival
for all donor/receptor/virus isolate combinations.

Numbers in the same column followed by different letters are
statistically different by Duncan's test (P < 0.05).









Table 3.2 Transmission of citrus tristeza virus by graft inoculation between selected citrus hosts: II. Overall rate of transmission.



Receptor host
Donor host Madam Vinous Mexican lime Grapefruit Average Citrus excelsa 72.41/2/b3/ 86.9 ab 60.7 b 72.54/b Mexican lime 93.1 a 76.9 b 89.3 a 85.2 ab Madam Vinous 86.7 ab 100.0 a 84.6 a 90.6 a
0I

1/ Percent transmission to plants with at least one inoculum piece (of four) alive 21 days
post-inoculation. Number indicates overall of transmission for all virus isolates
combinations.

2/ Each value represents a minimum of 27 plants. 3/ Numbers in the same column following by different letters are statistically different
by Duncan's tests (P 0.05).

4/ Number indicates the overall transmission for all receptor/virus isolate combinations.








47

differences (P : 0.05%) for all donor-receptor host combinations. Likewise, according to Duncan's multiple range comparison test, there were statistical differences between some of the hosts tested (Table 3.2).

The rate of transmission for the three different CTV isolates tested with each donor host is shown in Table 3.3. The T26 isolate was transmitted at a rate of 69.0% to 92.8%. The transmission rates of T30 and T66a isolates ranged from 62.5% to 100% and from 71.4% to 96.7%, respectively, from the hosts tested. While there were statistical differences in the rates of transmission for some of the virus isolate/donor host combination, the overall average of transmission showed no significant differences among them (Table 3.3).

The overall statistical analysis for percent transmission of the interactions among the different donor/receptor/virus isolate/inoculum pieces combinations, indicated no significant differences for receptor and virus isolates alone, and for the combinations of donor/inoculum pieces, receptor/inoculum pieces, and for donor/receptor/virus isolate. However, significant differences (P s 0.05) were found for donor and inoculum pieces alone, and for the interactions between donor/receptor, donor/virus isolate, receptor/virus isolate, and virus isolate/inoculum pieces.

Virus distribution and antigen concentration in host tissues. The relative antigen titer of CTV as measured by DAS-ELISA in each host tissue/virus isolate combination is









Table 3.3 Transmission of citrus tristeza virus by graft inoculation between selected citrus hosts: III. Effect of virus isolates.



Donor host
Virus Citrus Mexican lime Madam Vinous isolate excelsa sweet orange Average

T26 69.01/2/a3/ 89.3 a 92.8 a 83.54/a T30 76.7 a 65.2 b 100.0 a 80.7 a T66a 71.4 a 96.7 a 77.3 b 83.5 a


1/ Percent transmission to plants with at least one inoculum piece alive 21 days postinoculation. Number indicates overall of transmission for all donor/receptor host
combinations.

Each value represents a minimum of 27 plants.

/ Numbers in the same column following by different letters are statistically different
by Duncan's tests (P 5 0.05).

4/ Number indicates overall transmission for all donor/receptor host combinations.








49

illustrated in Table 3.4. The average optical density values at 405 nm (OD405) for bark tissue were 0.221 and 0.349 for the T26 isolate and 0.266 and 0.336 for the T66a isolate in Madam Vinous and C. excelsa, respectively. The OD405 values found in the other tissues assayed in both hosts for T26 and T66a isolates were in the range of 0.137 and 0.188 and 0.099 and 0.238 for petioles, 0.173 and 0.241 and 0.049 and 0.133 for midribs, and 0.044 and 0.065 and 0.030 for leaf blades, respectively. There were significant statistical differences between C. excelsa and Madam Vinous for bark tissue with the T26 isolate, and for both petioles and midribs with the T66a isolate. The overall analysis showed that the highest OD405 values in both hosts for both T26 and T66a isolates, were found in bark, followed by petioles and midribs. Leaf blades showed the lowest OD405 values of all tissues assayed in both hosts and isolates tested. Some differences in the OD405 values were found between different parts of the same plant, and from one plant to another, in some virus isolate/host combinations; however, the statistical analysis did not show significative differences among them (data not shown). From the standard curve prepared with purified T26 (Fig. 3.1), it was estimated that an OD405 value of 0.465 was approximately equivalent to 20 Ag/ml of CTV, assuming an extinction coefficient of 2.0 (Gonsalves et al. 1978). Therefore, the CTV antigen concentration in the test samples (10 mg of









Table 3.4 Relative antigen titer of citrus tristeza virus in different tissues of Citrus excelsa and Madam Vinous sweet orange host plants, as measured by enzyme-linked immunosorbent assay.



Host tissue

Virus Donor host Bark Petioles Midribs Leaf isolate blade T26 Citrus excelsa 0.3491/a2l 0.137 a 0.086 a 0.044 a

Madam Vinous 0.221 b 0.188 a 0.120 a 0.065 a
o

T66a Citrus excelsa 0.336 a 0.238 a 0.133 a 0.030 a

Madam Vinous 0.266 a 0.099 b 0.040 b 0.029 a



1/ Mean of optical density (OD405) per 10 mg plant tissue after 120 min of substrate reaction.
There were four replicates per plant and four plants per host/isolate combination.
Control reaction with the same tissue from healthy plants averaged OD405 0.001-0.025.
This has not been subtracted from the values above.

21 Numbers in the same column per host/isolate combination followed by different letters
are statistically different by Duncan's test (P < 0.05).









51




0.5 m 20

0.4 0
O
15
>12
0.3
U) o
0
0 10 0
0.2

.
C. 0.1 5
0 cc.

0 0 0 0.01 0.02 0.03 0.04 0.05
Purified virus (Optical Density 260 nm)






Figure 3.1 Plot of purified citrus tristeza virus (CTV) against optical density. Bark of healthy Citrus excelsa (0.25 g) tissue was ground in 5.0 ml of phosphate buffered saline, pH 7.6, + 0.05% Tween + 2% polyvinyl pyrrolidone and mixed with purified CTV T26 isolate to give the desired optical density at 260 nm (OD26). DAS-ELISA was performed as described in materials and methods. An extinction coefficient of 2.0 was assumed (Gonsalves et al. 1978) to estimate the relative virus concentration.








52

tissue/200 pl) ranged from an average of 0.2-0.5 Ag in leaf blades to 1.9-2.9 pg in bark tissue.

Discussion

In this study three CTV isolates were graft-transmitted by using leaf or bark tissue from three citrus donor hosts to three receptor hosts. The simple establishment and survival of grafted tissue in the receptor host was not sufficient to transmit CTV from certain donor hosts. There were a number of instances, 22 of 80, and 12 of 81, respectively, when C. excelsa and Mexican lime were used as donor hosts, where no transmission was achieved even on those receptor plants where at least one grafted tissue piece was still alive 21 days post-inoculation. Similar results were obtained, but to a lesser degree (7 of 75) when Madam Vinous sweet orange was the donor host. Furthermore, some of the receptor plants where no transmission was scored had all four grafted pieces still alive even five months post-inoculation.

The overall analysis of the results showed significant differences in the efficiency of the three donor hosts tested to transmit CTV (Table 3.2). Likewise, differences were found in the rate of transmission for each donor/receptor host combination. For example, C. excelsa showed rates of transmission of 72.4% and 60.7% to Madam Vinous and grapefruit, respectively; whereas, a rate of transmission of 86.7% was obtained to Mexican lime plants. In regard to Mexican lime as donor host, there was a rate of 89.3% and








53

93.1.7% transmission to grapefruit and Madam Vinous, respectively and a 76.9% rate to Mexican lime. Transmission from Madam Vinous sweet orange was between 84.6 and 100% in all receptors tested. It was surprising that transmission rates between the same species were only 86.7% for Madam Vinous, and 76.9% for Mexican lime (Table 3.2).

Madam Vinous sweet orange was the most efficient donor host with the three receptor hosts tested (90.6%), followed by Mexican lime (85.2%). C. excelsa was a poor donor host (72.5%), being relatively efficient only when inoculated to Mexican lime (Table 3.2).

Previous studies on the transmission of CTV by grafting procedures have shown that a period of at least ten days contact between grafted tissues is needed to obtain transmission of the virus to the receptor host (Tolba et al. 1976; Yamaguchi and Patpong, 1980). In this study, the survival of grafted tissue was scored 21 days after inoculation, but the inoculated tissue was left in the receptor plants for up to five months. This should have been ample time for contact between the inoculum and receptor cambium to establish a tissue union with a subsequent transmission of CTV. Furthermore, when leaf pieces were used as inoculum, a small portion of the midrib was included in every piece to increase the success of the grafting. The overall rates of successful grafts were about 83% and a 66.7% for leaf and bark pieces, respectively (Table 3.1).








54

The reason why a low percentage of graft transmission of the virus was found from some donor hosts, and the absence of an expected 100% when the donor-receptor combination was of the same species, is unknown. A possible explanation could be differences in the virus distribution and/or concentration in the donor tissues used as inoculum. Bark tissue contained the highest antigen titer with OD405 values in the range of 0.221 and 0.349 in both C. excelsa and Madam Vinous with both CTV isolates tested (Table 3.4). These values were, in some instances, more than double those found in petioles and midribs, and at least triple those found in the leaf blade. Even though the statistical analysis did not show significant differences in antigen titer in either different parts of the same plant or from one plant to another, there were some instances where OD405 values were as low as the healthy controls. This indicates a possible absence of the virus in those tissues and raises the possibility that occasionally the tissue used for graft transmission may be virus-free, with a subsequent failure in the transmission. Other possibilities could be an occasional absence of phloem connections between the donor and receptor tissues with a subsequent absence of movement of the virus across the junction or the requirement of a minimum of virus particles present in the tissue used as inoculum in order to accomplish the transmission.

Citrus tristeza virus is phloem-limited (Bar-Joseph et al. 1979a; Lister and Bar-Joseph, 1981), and is normally found








55

at higher concentrations in young phloem-rich tissues (Garnsey, et al. 1979; Bar-Joseph, et al. 1979a); however, the serological titer frequently decreases as the tissues reach maturity or when the plants are exposed to warm environments (Garnsey et al. 1981a; Lee et al. 1988c). The OD405 values obtained in this research were low if compared with those found when DAS-ELISA is used routinely for CTV diagnosis (Bar-Joseph et al. 1979b; Garnsey et al. 1980b); however, this part of the work was addressed to determine the virus titer in the tissues suitable for graft transmission, and young tender tissue sometimes is not a good source of inoculum for leaf piece grafts (personal observations).

The overall analysis of the results obtained indicates that the efficiency of the graft transmission of CTV is conditioned primarily by the donor/receptor host combination, and secondly by the virus isolate involved, but apparently not by the interaction of the three. For example, C. excelsa showed an overall rate of transmission in the range of 72.5% with all receptor hosts tested (Table 3.2), and a similar low pattern between 69% and 76.7% (= 72.8%) was obtained for the three isolates tested (Table 3.3). Likewise, when Madam Vinous was used as the donor host, there was an overall rate of transmission of 90.6% (Table 3.2). A rate of 77.3-100% (= 88.6%) occurred from this host with the three isolates tested (Table 3). A comparable event was also scored when Mexican lime was the donor host (Table 3.2 and 3.3). The statistical








56

significance found for the interactions donor/receptor and donor/virus isolate, and no significance for the interaction of donor/receptor/virus isolate supports this conclusion.

The use of leaf and/or bark pieces for graft transmission of CTV may be advantageous when large numbers of plants are to be inoculated with limited sources of inoculum (Garnsey and Whidden, 1970). However, in the light of the results of this research, in order to achieve a high level of transmission, the efficiency of the donor host and the donor/receptor host combination should be considered.














CHAPTER 4
DEVELOPMENT OF A DOT-IMMUNOBINDING ASSAY FOR CITRUS TRISTEZA VIRUS


Introduction

Several serological methods have been developed and used to diagnose the presence of citrus tristeza virus (CTV) in infected tissue. These methods include: SDS-immunodiffusion procedures (Garnsey et al. 1979; Bar-Joseph et al. 1980), in situ immunofluorescence (Brlansky et al. 1984; Tsuchizaki et al. 1978), serologically specific electron microscopy (Brlansky et al. 1984; Garnsey et al. 1980), gold immunolabeling microscopy (Davis and Brlansky, 1990), and several enzyme-linked immunosorbent (ELISA) procedures (BarJoseph and Malkinson, 1980; Bar-Joseph et al. 1979b) including the use of the biotin-avidin system (Irey et al. 1988) and the enzyme-amplified ELISA (Ben-Ze'ev et al. 1988) to increase the sensitivity of detection. Each of these methods has different advantages and sensitivity levels, and therefore, has been used for different purposes and applications (Rocha-Pefa and Lee, 1990).

Polyclonal antisera specific for CTV have been developed in different animal species, such as rabbits (Gonsalves et al. 1978; Tsuchizaki et al. 1978; R.F. Lee, unpublished;), and


57








58

chickens (Bar-Joseph and Malkinson, 1980; Marco and Gumpf, 1990). Likewise, several monoclonal antibodies (MCAs) have been developed in mice (Gumpf et al. 1987; Permar et al. 1990; Vela et al. 1986, 1988). The 3DF1 MCA has been reported to react with a broad spectrum of CTV isolates of different geographical origins and is available commercially (Vela et al. 1986, 1988). The MCA-13 MCA (hereafter MCA-13) has been reported to react specifically with CTV isolates that have severe biological activities, especially isolates causing decline on plants grafted on sour orange rootstock (Garnsey and Permar, 1990; Permar et al. 1990). It has been used in diverse studies for strain discrimination purposes (Irey et al. 1988; Garnsey and Permar 1990; Rocha-Peia et al. 1990; Yokomi et al. 1990).

The ELISA test, particularly the double antibody sandwich system (DAS-ELISA) (Bar-Joseph et al. 1979b; 1980) has been the most widely used of all serological methods developed for CTV detection (Bar-Joseph et al. 1989; Garnsey, et al. 1981a; Rocha-Pefa and Lee, 1991). In DAS-ELISA the virus in the test sample is trapped and immobilized selectively by specific antibodies adsorbed on polystyrene microtiter plates. Enzymeconjugated antibodies are then reacted with the trapped virus and detected colorimetrically after adding a suitable substrate (Clark and Adams, 1977; Garnsey and Cambra, 1990). DAS-ELISA is relatively easy to perform and is highly sensitive, but it does require some special equipment, it is








59

laborious and time-consuming for large scale indexing, and both antibodies and large volumes of buffer are used in each test.

A simple and rapid serological method, known as dotimmunobinding assay (DIBA) (Hawkes et al. 1982), has been developed and applied for the detection of several plant viruses (Hibi and Saito, 1985; Powell, 1987). The principles of DIBA are similar to those of ELISA, differing mostly in that the antigen and antibodies are bound to nitrocellulose membranes instead of polystyrene microtiter plates, and that the product of the enzyme reaction at the end of the test is insoluble. The use of nitrocellulose membranes for the serological detection of plant viruses has become popular because of the simplicity of equipment required and lower cost of materials needed. In a preliminary study the DIBA test was as sensitive as DAS-ELISA to detect CTV in both field trees and greenhouse grown plants (Rocha-Pefa et al. 1990); however, some differences were found in the reactivity of the antibodies used, and some nonspecific reactions occurred in the test. The objective of this research was to adapt DIBA for diagnosis of CTV using different polyclonal and monoclonal antibodies and to compare the sensitivity of DIBA with DASELISA and DAS-indirect ELISA.

Materials and Methods

Antisera used. Polyclonal antisera numbers 1051, 1052, and 1053 previously prepared in rabbits against undegraded,








60

and unfixed virus particles of T30, T36, and T26 CTV isolates, respectively (R.F. Lee, unpublished) were used.

Immunoglobulins (IgG) were purified from whole sera by the Protein A-Sepharose affinity chromatography method (Miller and Stone, 1978) and adjusted to a final concentration of 1.0 mg/ml (OD2,= 1.40) in PBS buffer with 0.02% sodium azide and stored at 40C (Clark et al. 1986). The 3DF1 MCA was a gift from Drs. P. Moreno and M. Cambra, Valencia, Spain, and its preparation was described previously (Vela et al. 1986, 1988). The MCA-13 that reacts specifically with severe CTV strains (Permar et al. 1990) was a gift from Drs. T.A. Permar and S.M. Garnsey. Goat anti-mouse and goat anti-rabbit IgG conjugated with alkaline phosphatase were purchased from either Boehringer or Promega.

The purified polyclonal IgG and 3DFI MCA were tested at concentrations of 1.0, 0.1, 0.2 or 0.01 Ag/ml. The MCA-13 was used as ascites fluid at a dilution of 1:5,000 (v/v). Goat anti-species IgG were used at concentrations recommended by the manufacturer.

Sample preparation.- Bark tissue was peeled from fullyexpanded new flushes of CTV infected and healthy citrus plants. The tissue was finely chopped and homogenized with a Tekmar Tissumizer in the presence of Tris buffered saline (TBS)-Tween (TBS = 0.02 M Tris, 0.5 M NaCl, pH 7.5), plus 0.5 % Tween 20 (TBS-Tween) at 1:10 (w/v) dilution.








61

Dot-immunobindinQ assay (DIBA).- The general protocol used for DIBA was as follows: nitrocellulose membranes (Micro Separations, Inc.), 0.45 A pore, were cut at a size of 11 X 7.5 cm and wet in TBS for at least 30 min, blotted on chromatography paper (Whatman No. 1), and allowed to dry for 5 min before use. Aliquots of 2 Al of test samples were applied to nitrocellulose membranes by using as a guide a template constructed from the rack of a micropipet tip holder box, allowed to dry for at least 10-15 min or stored at room temperature for several days before use. All subsequent incubation steps were performed at room temperature in 25 ml of each solution using polypropylene covers of micropipet tip holder boxes as trays. During the incubation or washing the membranes were agitated gently in a shaker at 50 oscillations per min or agitated by hand. Nitrocellulose membranes, with the test samples, were soaked for 30 min in blocking solutions of either 10% horse serum (v//v), 3% bovine serum albumin (BSA) (w/v), 3% gelatin (w/v), 5% Triton X-100 (v/v), or 0.5% non-fat dry milk (w/v), all in TBS, and washed twice with 2550 ml TBS-Tween and once with TBS, 2 min each. Then the membranes were incubated with the CTV specific antibodies for either 30 min, one, two, or 18 hr, depending on the IgG used, and washed again as after blocking. The membranes were then incubated for 30 min with the corresponding goat anti-species IgG conjugated with alkaline phosphatase. After washing as before, the membranes were incubated in the substrate








62

solution. The substrate solution was prepared as follows: 10 mg of nitro blue tetrazolium (NBT) (Sigma) were dissolved in 30 ml of TBS substrate buffer (0.1 M Tris, 0.1 M NaC1l, 0.005 M MgCl2, pH 9.5); then 5 mg of 5-bromo-4-chloro-3-indoyl phosphate (BCIP) (Sigma) were added and dissolved in the solution. The color reaction was stopped by transferring the membranes to distilled water.

DAS-ELISA.- The double antibody sandwich enzyme-linked immunosorbent assay (DAS-ELISA) (Bar-Joseph et al. 1979b, 1980) was conducted using polyclonal IgG No. 1053 and polystyrene Immulon II microtiter plates (Dynatech Laboratories). Unless otherwise stated, 200 microliters were used per well and three washings with PBS-Tween (phosphate buffered saline = 8 mM Na2HPO4, 14 mM KH2PO4, 15 mM NaCl, pH 7.4, + 0.05% Tween 20) were performed between steps. Microtiter plates were coated using 3.0 Ag/ml of purified CTV specific IgG in carbonate buffer (0.015 M NaHCO3, 0.03 M NaCO3, pH 9.6) and incubated for 6 hr at 370C. Antigen samples were added to the wells and incubated for 18 hr at 50C. CTV specific IgG conjugated with alkaline phosphatase at a dilution of 1:1,000 in conjugate buffer {PBS-Tween + 2% polyvinyl pyrrolidone (=PVP 40,000 MW) (w/v) + 0.2% bovine serum albumin (w/v)} was incubated for 4 hr at 370C. The reaction with p-nitrophenyl phosphate (Sigma) (1.0 mg/ml in 10% triethanolamine, pH 9.8) was quantified after 60 min at








63

405 nm (OD4~) using a Labinstruments model EAR 400 AT ELISA plate spectrophotometer.

DAS-indirect ELISA.- For DAS-Indirect ELISA the plates were coated with polyclonal IgG No. 1053, and antigen samples were added as described for DAS-ELISA. The 3DF1 MCA was used at a concentration of 0.2 Ag/ml IgG and MCA-13 ascites fluid was used at a dilution of 1:5,000 (v/v). Each was diluted in conjugate buffer, added to the appropriate plates and incubated for 4 hr 370C. After washing, alkaline phosphatase conjugated goat anti-mouse IgG was added at a dilution of 1:7,500 (v/v) in conjugate buffer and incubated for 2 hr at 37C. The enzyme-substrate reaction was carried out as described for DAS-ELISA.

Evaluation.- Twelve selected naturally-occurring Florida CTV isolates with different biological properties (Garnsey et al. 1987; Permar et al. 1990; Rocha-Pefia and Lee, unpublished) were used to determine the range of reactivity of each antibody. All virus source plants were maintained in a greenhouse with mean minimum and maximum temperatures of 21 and 380C, respectively. The isolates T11a, T26, T30, T50a and T55a produce very mild symptoms and little or no stunting in Mexican lime seedlings {Citrus aurantifolia (Christm.) Swingle}, no seedling yellows on sour orange (C. aurantium L.) or grapefruit (C. paradisi Macf.) and no symptoms in sweet orange {C. sinensis (L.) Osb.} and sweet/sour orange combinations. The isolate T4 causes a moderate or stronger








64

reaction on Mexican lime than isolates listed above. The isolates T3, T36, T62a, T65a, T66a and T67a cause strong vein clearing, stunting and stem pitting in Mexican lime seedlings, varying degrees of decline and/or growth reduction in sweet/sour orange combinations, varying degrees of seedling yellow reaction and/or severe stunting in sour orange or grapefruit seedlings, and varying degrees of growth reduction in Madam Vinous sweet orange seedlings.

The relative sensitivity limits were compared for detection of CTV by DIBA, DAS-ELISA, and DAS-indirect ELISA. An extract {1:10 (w/v) in TBS-Tween} was made of greenhouse grown C. excelsa plants infected with the CTV isolate T36. A series of two fold dilutions was made with a similar extract of healthy C. excelsa. Negative controls included TBS-Tween, and 1:10 extracts of healthy g. excelsa, Madam Vinous sweet orange, Duncan grapefruit and Mexican lime plants, all in TBSTween.

Evaluation of different buffers for sample extraction.An additional experiment was carried out to determine the effects of different extraction buffers on the sensitivity of DIBA, DAS-ELISA, and DAS-indirect ELISA. Three different extraction solutions, Tris buffered saline, phosphate buffered saline, and carbonate buffer (all as described above), with and without 0.05% Tween 20, were used. Three CTV isolates T26, T62a, and T66a propagated in Madam Vinous sweet orange plants were tested in DIBA, DAS-ELISA, and DAS-indirect ELISA








65

with polyclonal IgG No. 1053 and the 3DF1 MCA as described before. Polyclonal IgG No. 1053 (1.0 Ag/ml) was incubated for 2 hr at 370C in 1% BSA, 2% PVP, TBS containing 1:200 (v/v) buffer extract of bark from healthy Madam Vinous sweet orange, prior to use in DIBA. Substrate reaction for DAS-ELISA and DAS-indirect ELISA was quantified after 60 min.

Results

Development of the dot-immunobinding assay. The reactivity of each antibody and level of background on the nitrocellulose membranes varied with each IgG used, IgG concentration and incubation time. Reactions were affected by the blocking solutions and buffers used for dilution of the antibodies and the commercial goat anti-species IgG conjugates.

Six different solutions were evaluated as blocking agents. TBS alone (Fig. 4.1 A) and 10% horse serum (not shown) did not prevent the nitrocellulose sheets from turning dark; whereas 3% BSA, 3% gelatin, 0.5% non-fat dry milk, and 5% Triton X-100 all gave an acceptably white membrane with the different polyclonal IgGs tested (Fig 4.1 B, C, D, E). The 3% gelatin blocking solution gave the best contrast between the green color for healthy samples and different intensities of a purple color for CTV infected samples (Fig. 4.1 C). The use of Triton X-100 as a blocking agent partially removed the green material from the nitrocellulose sheets; however, a








66





























Fig. 4.1 Effect of different blocking solutions on the reaction of polyclonal antibodies no. 1053 in dotimmunobinding assay with citrus tristeza virus (CTV) isolates: A. TBS alone; B. 3% bovine serum albumin (BSA); C. 3% gelatin; D. 0.5% non-fat dry milk; E. 5% Triton X-100. Number 1, CTV T66a; 2, CTV T26; 3, buffer extract from healthy sweet orange plants; 4, TBS-Tween. Reaction conditions are described in materials and methods.








67

slight pink reaction was consistently obtained with sap from healthy plants (Fig. 4.1 E).

Polyclonal IgGs had to be cross absorbed with buffer extracts of healthy tissue to reliably discriminate between healthy and CTV infected samples (Fig. 4.2 A, B), and reactivity varied with concentration and incubation times. The strongest reactions and best discrimination between CTV positive and negative samples were achieved using 1.0 pg IgG/ml and 30 min incubation. An incubation time of 60 min with 1.0 or 0.1 Ag IgG/ml, or longer (18 hr) even with as low as 0.01 Ag IgG/ml frequently resulted in the occurrence of nonspecific reactions with healthy samples, even when the IgG had been cross absorbed with extracts of healthy plants. sA would be expected, MCAs did not have to be cross absorbed with buffer extracts from healthy plants prior to use. The reactivity of the 3DF1 and MCA-13 MCAs was substantially lower than that obtained with any of the polyclonal IgGs tested. For both 3DF1 (1.0 Ag/ml) and MCA-13 (diluted 1:5,000), the incubation time was extended to 2 hr or longer (18 hr) to achieve an adequate positive reaction with infected samples (Fig. 4.2 C, D). The reactivity of 3DF1 was always slightly lower than that obtained with either polyclonal IgG or the MCA-13.

In initial experiments with polyclonal IgGs, 1% BSA in TBS plus 2% PVP was chosen as the antibody diluent because it consistently gave a whiter background on the nitrocellulose








68





























Fig. 4.2 Reaction of polyclonal and monoclonal antibodies in dot-immunobinding assay with citrus tristeza virus (CTV) isolates. A. Polyclonal antibodies 1053 (1.0 Ag/ml) not cross-absorbed with buffer extract of healthy plant. B. Polyclonal antibodies 1053 (1.0 Ag/ml) incubated with 1:200 (v/v) buffer extract from healthy plants for 2 hr at 370C prior to use. C and D, monoclonal 3DF1 (1.0 Ag/ml) and MCA13 (1:5,000 dilution) antibodies not cross-absorbed with healthy extract. Number 1, CTV T66a; 2, CTV T26; 3, buffer extract from healthy sweet orange plants; 4, TBS-Tween. Reaction conditions are described in materials and methods.








69

and stronger reactivity for positive samples. Either the absence of PVP or BSA concentrations of less than 1% in the antibody diluent frequently resulted in a dark background on the nitrocellulose when 3% BSA was used for blocking.

Evaluation.- The relative sensitivity level of DIBA was measured using increasing dilutions of a plant extract containing CTV isolate T36. This is shown in Figure 4.3. Polyclonal IgGs Nos. 1051, 1052, and 1053 (1.0 gg/ml IgG for 30 min) consistently gave a distinct positive reaction with CTV T36 diluted to 1/160. There was a slight but consistently positive reaction at the 1/320 dilution and occasionally at 1/640 (Fig. 4.3 A, B, C). The reactivity of both 3DF1 and MCA-13 MCAs (1.0 Ag/ml and 1:5,000, respectively) was also in the range of 1/160 and 1/320, but only when the incubation time was extended to 18 hr (Fig. 4.3 D, E).

The relative sensitivities of DAS-ELISA and DAS-indirect ELISA were measured similarly against the diluted extract of CTV T36. These results are summarized in Table 4.1. DASELISA, using polyclonal IgG No. 1053 for both coating and conjugate steps, gave OD405 value of 0.109 (i.e above 0.100) with the CTV T36 sample diluted to 1/160. This was considered a positive reaction. Thus the OD405 value of 0.070 at a 1/320 dilution was considered a negative reaction. DAS-indirect ELISA, using polyclonal IgG No. 1053 for coating and 3DF1 MCA as second antibody in the double sandwich, gave a positive and negative reaction at dilutions of 1/320 and 1/640,








70






A B C D E 1/10 *

1/20 1/40 1/80

1/160

1/320

1/640

TBS-Tween




Fig. 4.3 Relative sensitivity level of different polyclonal and monoclonal antibodies specific to citrus tristeza virus (CTV) in dot-immunobinding assay. Rows A, B and C, polyclonal antibodies nos. 1051, 1052, and 1053, respectively. Rows D and E, monoclonal antibodies 3DF1 and MCA-13, respectively. Extract of Citrus excelsa greenhouse grown plants infected with CTV T36 isolate was prepared in TBS-Tween 1:10 (w/v) and two-fold diluted with extract of healthy plants. Reaction conditions are described in materials and methods.








Table 4.1 Relative sensitivity of DIBA, DAS-ELISA and DAS-indirect ELISA with polyclonal (1053) and monoclonal (3DF1 and MCA-13) antibodies specific to citrus tristeza virus.


CTV T-3. DIBA DAS-ELISA1/ DAS-indirect ELISA1/ isolate-/
1053 3DF1 MCA-13 1053 3DF1 MCA-13 1/10 +31 + + 0.572 +4/ 2.628 + 0.892 + 1/20 + + + 0.460 + 2.507 + 0.607 + 1/40 + + + 0.314 + 1.635 + 0.385 + 1/80 + + + 0.185 + 0.853 + 0.252 + 1/160 + + + 0.109 + 0.518 + 0.158 + 1/320 + +/- +/- 0.070 - 0.190 + 0.094 1/640 - - - 0.024 - 0.075 - 0.054 1/1280 - - 0.022 - 0.032 - 0.036 C. excelsa - - 0.004 - 0.033 - 0.013


1/ The IgG of polyclonal antibody no. 1053 was used to coat the plates for both DASELISA and DAS-indirect ELISA. For DAS-ELISA the no. 1053 IgG conjugate was used as
second antibody. For DAS-indirect ELISA the unlabeled 3DF1 and MCA-13 were the
intermediate antibodies followed by the goat anti-mouse IgG conjugate.









2/ Buffer extract (1:10 dilution) of bark from greenhouse grown C. excelsa plants
infected with CTV T36 successively two-fold diluted with healthy C. excelsa buffer
extract.

3/ Visual evaluation for presence of a purple color (+ = positive, +/- = inconclusive,
= negative.
4/ Optical density at 405 nm (OD405) after 60 min of substrate reaction. Mean of two
replicatiqns per plate. Reactions were considered positive (+) when OD405 values were higher than three times the mean of healthy controls or 0.100, whichever was greater.
Reactions with lower values were considered negative (-).








73

respectively. DAS-indirect ELISA using the MCA-13 as second antibody had a sensitivity level similar to DAS-ELISA , i.e. the dilution end point was 1/160 dilution (Table 4.1).

The relative reactivity of the different antibodies for the 12 selected CTV isolates in DIBA, and its comparison with both DAS-ELISA and DAS-indirect ELISA, is illustrated in Figure 4.4 and Table 4.2. Polyclonal IgG No. 1053 (1.0 gg/ml) showed a strong positive reaction with all CTV isolates tested in DIBA (Fig. 4.4, left). Polyclonal IgGs Nos. 1051 and 1052 reacted similarly (not shown). The results obtained with polyclonal IgG No.1053 in DAS-ELISA with the CTV isolates were the same as with DIBA (Table 4.2).

The 3DF1 MCA (1.0 Ag/ml IgG) reacted moderately with most CTV isolates tested, but no reaction and weak or inconclusive reactions were obtained with the isolates T26 and T66a, respectively, in both DIBA and DAS-indirect ELISA (Fig. 4.4, center and Table 4.2).

The severe strain specific MCA-13 (used at a dilution of 1:5,000) gave a distinctly positive reaction only with isolates T3, T36, T65a, T66a, and T67a in DIBA and DASindirect ELISA (Fig. 4.4, right and Table 4.2). An inconclusive or slightly positive reaction was obtained occasionally with the T50a, T55a, T4, and T62a isolates in DIBA when the samples were incubated for longer times (18 hr). Similar reactions occurred with these particular CTV isolates in the DAS-indirect ELISA test (Table 4.2).








74





AB A B AB

1 u * 2 � 3 a

4 g 0)

5 *



7 Q





1053 3DF1 MCA-13,



Fig. 4.4 Reaction of polyclonal antibodies no. 1053 and 3DF1 and MCA-13 monoclonal antibodies in dot-immunobinding assay with twelve selected citrus tristeza virus (CTV) isolates. Row A: 1 = T11a; 2 = T26; 3 = T30; 4 = T50a; 5 = T55a; 6 = T3; 7 = T4; 8 = T36. Row B: 1 = T62a; 2 = T65a; 3 = T66a; 4 = T67a. Also in Row B are extract of healthy plants: 5 = Citrus excelsa; 6 = Madam Vinous sweet orange; 7 = Duncan grapefruit; 8 = Mexican lime. Reaction conditions are described in materials and methods.








Table 4.2 Comparison of DIBA, DAS-ELISA and DAS-indirect ELISA for the detection of citrus tristeza virus (CTV) and relative reactivity of polyclonal (1053) and monoclonal (3DF1 and MCA-13) antibodies with CTV isolates.


Virus DIBA DAS-ELISA1/ DAS-indirect ELISA1/ isolated/
1053 3DF1 MCA-13 1053 3DF1 MCA-13 Tlla (M)/ +4/ + - 0.204 +5/ 1.568 + 0.091 T26 (M) + +/- - 0.137 + 0.053 - 0.057 T30 (M) + + 0.227 + 1.936 + 0.090 T50a (M) + + +/- 0.498 + 2.584 + 0.120 + T55a (M) + + +/- 0.443 + 2.018 + 0.119 + T3 (S) + + + 0.404 + 2.468 + 0.522 + T4 (MD) + + +/- 0.304 + 1.992 + 0.092 T36 (S) + + + 0.572 + 2.628 + 0.892 + T62a (S) + + +/- 0.556 + 2.045 + 0.115 + T65a (S) + + + 0.556 + 2.527 + 0.745 + T66a (S) + +/- + 0.588 + 0.191 + 0.735 + T67a (S) + + + 0.250 + 2.235 + 0.480 +










C. excelsa (HC) - 0.021 - 0.033 - 0.013

M. Vinous (HC) - 0.004 - 0.045 - 0.000

Grapefruit (HC) - 0.004 - 0.014 - 0.008

M. lime (HC) - 0.009 - 0.029 - 0.038



1/ The IgG of polyclonal antibody no. 1053 was used to coat the plates for both DASELISA and DAS-indirect ELISA. For DAS-ELISA the no. 1053 IgG conjugate was used as
second antibody. For DAS-indirect ELISA the unlabeled 3DF1 and MCA-13 were the
intermediate antibodies followed by the goat anti-mouse IgG conjugate.

2/ Buffer extracts (1:10 dilution) of bark from greenhouse grown plants either of C.
excelsa, Madam Vinous sweet orange, grapefruit or Mexican lime.

/ M = mild; MD = moderate; S = severe; HC = healthy control; on the basis of symptom
reaction on a series of citrus hosts (Garnsey et al. 1987; Permar et al. 1990; RochaPefa and Lee, unpublished).

4/ Visual evaluation for presence of a purple color (+ = positive, +/- = inconclusive,
= negative).

5/ Optical density at 405 nm (ODA05) after 60 min of substrate reaction. Mean of two
replications per plate. React ons were considered positive (+) when OD405 values were higher than three times the mean of healthy controls or 0.100, whichever was greater.
Reactions with lower values were considered negative (-).








77

Evaluation of different buffers for sample extraction.The reactions of the polyclonal IgG 1053 and MCA 3DF1 with the different CTV isolate/extraction buffer combinations in DIBA are shown in Figure 4.5. In general, there was a strong positive reaction with all the CTV isolate/extraction buffer combinations tested, with the strongest reaction occurring when PBS without Tween 20 was used as the extraction buffer (Fig. 4.5). The presence of Tween 20 in the extraction buffer gave a weaker but relatively more uniform reaction spot with both antibodies.

The results of the evaluation of the different extraction buffers by DAS-ELISA and DAS-indirect ELISA are shown in Figure 4.6. In DAS-ELISA the strongest reactions occurred with isolate T62a, followed by T26 and T66a, respectively. The presence or absence of Tween 20 made little difference in DAS-ELISA (Fig. 4.6 left). In DAS-indirect ELISA with the 3DF1 MCA the strongest reactions occurred with isolate T26, followed by T62a and T66a, respectively. The presence of Tween 20 gave slightly stronger reactions in most cases. With PBS, the presence of Tween 20 more than doubled the OD405 values for isolates T62a and T66a, but caused only a slight increase for isolate T26 (Fig. 4.6 right) Discussion

The dot-immunobinding assay was adapted for detection of CTV. This included the testing of different agents for blocking and as diluents for the CTV specific antibodies and









78






















containing 0.05% Tween 20, PBS = phosphate buffered saline,
M *0. 0e.

CbCarb + 0.05% Tween, HC = healthy control. The IgG of











polyclonal antibodies No. 1053 or monoclonal antibody 3DF3DF1
Figure 4.5 Evaluation of different extraction buffers on the sensitivity of DIBA with citrus tristeza virus (CTV) isolates T26, T62a and T66a. TBS = Tris buffered saline, TBST = TBS containing 0.05% Tween 20, PBS = phosphate buffered saline, PBST = PBS + 0.05% Tween, Carb = carbonate buffer, CarbT = Carb + 0.05% Tween, HC = healthy control. The IgG of polyclonal antibodies No. 1053 or monoclonal antibody 3DF1 were used as primary antibodies followed by the goat antirabbit and anti-mouse IgG conjugates, respectively. Samples were 2 Al of buffer extracts of greenhouse-grown Madam Vinous sweet orange plants infected with the CTV isolates ground at a 1:10 dilution.















-.d











DAS-ELISA DAS-indirect ELISA




S0.7 - o 0.7 S.6 0.6S 0.5- 0.5 0.4 0.4
0.3- 0.3
o 0.2- 0.2 a U1- C 0.10.1
0 0 0
T26 T62a T66a HC T26 T62a T66a HC

W eTBS TBST O PB TBS TBST - PBS
PBST E Carb CarbT PBST M Carb M CarbT




Figure 4.6 Evaluation of different extraction buffers on the sensitivity of DAS-ELISA and DAS-indirect ELISA with citrus tristeza virus (CTV) isolates T26, T62a and T66a.
TBS = Tris buffered saline, TBST = TBS containing 0.05% Tween 20, PBS = phosphate buffered saline, PBST = PBS + 0.05% Tween, Carb = carbonate buffer, CarbT = Carb + 0.05% Tween, HC = healthy control. The IgG of polyclonal antibodies No. 1053 was used to coat the plates for both DAS-ELISA and DAS-indirect ELISA. For DAS-ELISA the No. 1053 IgG conjugate was used as second antibody. For DAS-indirect ELISA the unlabeled 3DF1 monoclonal antibody was the intermediate antibody followed by the goat anti-mouse IgG conjugate. Samples were 200 Al of buffer extracts of greenhouse-grown Madam Vinous
sweet orange plants infected with the CTV isolates ground at a 1:10 dilution.








80

the commercial goat anti-species IgG conjugates, the evaluation of some CTV specific polyclonal and monoclonal antibodies and the testing of the effect of different extraction buffers with and without Tween 20.

In some other systems where nitrocellulose membranes have been used for the immunological detection of proteins either by DIBA (Hibi and Saito, 1985; Powell, 1987; Graddon and Randles, 1986) or by western blotting (Spinola and Cannon, 1985), some differences have been reported in the suitability of different agents and buffers used for membrane blocking or as diluents for both the virus specific antibodies and the commercial goat anti-species IgG conjugates. In this study, TBS, 10% horse serum, 3% BSA, 3% gelatin, 0.5% non-fat dry milk and 5% Triton X-100 were evaluated as blocking agents. The latter four were found to give an adequately white background on the nitrocellulose membranes to permit discrimination between infected and healthy samples. However, 3% gelatin was used routinely as it gave the most suitable contrast between a green color for the healthy samples and a distinct purple color for the infected samples.

It has been well documented that the presence of albumin (BSA or egg ovalbumin) enhances the reactivity of antibodies in diverse serological tests (Clark et al. 1986; Purcifull and Batchelor, 1977). Likewise, PVP has been commonly used in both extraction and conjugate buffers to prevent non-specific reactions (Clark et al. 1986). In this study, a concentration








81

of BSA of less than 1% or the absence of PVP in the antibody diluent frequently resulted in the nitrocellulose membrane turning dark when BSA was used for blocking. Thus TBS containing 1% BSA and 2% PVP was always used as the antibody diluent.

Differences were found in the reactivities of the antibodies tested. All polyclonal IgGs (Nos. 1051, 1052, and 1053) reacted similarly in DIBA. After 30 min incubation (at 1.0 Al/ml IgG) they gave strong positive reactions with CTVinfected samples, but none with healthy samples. However, incubations of 60 min or more sometimes resulted in the occurrence of nonspecific reactions with extracts of healthy plants. This was prevented when the polyclonal antibodies were cross-absorbed with extracts of healthy plants before use. These antibodies gave a strong positive reaction with the 12 selected CTV isolates tested, and had a sensitivity (highest dilution which gave positive reaction) limit of 1/320 with CTV T36 infected samples.

The 3DF1 MCA when used at an IgG concentration of 1.0 Ag/ml reacted moderately with most of the CTV isolates tested and had a relative sensitivity limit of 1/320 in DIBA and 1/680 in DAS-indirect ELISA. However, in some cases it was necessary to extend the incubation time (15-18 hr) with 3DF1 in DIBA, because it had a reactivity level that was somewhat lower than any of the polyclonal antibodies tested.








82

The 3DF1 MCA has been reported to react with a broad spectrum of CTV isolates primarily on the basis of DASindirect ELISA (Vela et al. 1986, 1988). In this work the 3DF1 reacted weakly with CTV isolates T26 and T66a in some tests (Fig. 4.4, center, A2 and B3) and moderately in others (Fig. 4.2, C1,2,5,6). The reason for this is not known. For isolate T26 it could be due to a low virus concentration as was indicated by DAS-ELISA (Table 4.2). However, the T66a isolate was at a high concentration as indicated in DAS-ELISA with polyclonal antibody 1053 and in DAS-indirect ELISA with MCA-13 (Table 4.2). Yet, in DAS-indirect ELISA 3DF1 gave a low OD405 value for T66a (Table 4.2). This may indicate a differential reactivity of the 3DF1 MCA with some particular CTV isolates such as T66a.

The sensitivity limit for MCA-13 in both DIBA and DASindirect ELISA was near 1/320 (Fig. 4.3 and Table 4.1). In DIBA MCA-13 reacted strongly with CTV isolates T3, T36, T65a, T66a, and T67a, all of which share severe biological properties (Garnsey et al. 1987; Permar, et al. 1990; RochaPefia and Lee, unpublished). However, there were slight positive reaction with CTV isolates T50a and T55a which have mild biological properties and with isolate T62a which has severe properties (Rocha-Peia and Lee, unpublished) (Table 4.2). The MCA-13 has been reported to react specifically with CTV isolates that have severe biological properties, such as decline, stem pitting, seedling yellows, etc. (Permar and








83

Garnsey, 1990; Permar et al. 1990), and it has been used for strain discrimination of CTV-induced decline in several field surveys (Irey, et al. 1988; Rocha-Pefa et al. 1990; Yokomi et al. 1990). The slight positive reaction found frequently in DIBA with some CTV isolates with mild biological properties occurred when the ascites fluid at a dilution of 1:5,000 was incubated for 18 hr. These slight positive reactions could be prevented by using a shorter (2 hr) incubation time or by using the ascites fluid at a dilution of 1:20,000 and an incubation time of 18 hr (Rocha-Pefia et al. 1990). The isolate T62a has been characterized previously as severe on the basis of a severe vein flecking, leaf cupping and stunting of Mexican lime plants, severe growth reduction on grapefruit and Madam Vinous sweet orange seedlings, and moderate growth reduction on sweet/orange combinations (Rocha-Pefia and Lee, unpublished). The weak reaction found with isolate T62a in DIBA and its low OD405 when MCA-13 was used in DAS-indirect ELISA (Table 4.2), suggests that some severe CTV isolates may not be recognized by the MCA-13. That T62a was in high concentration in these experiments is verified by DAS-ELISA using polyclonal antibodies and DAS-indirect ELISA using the 3DF1 MCA (Table 4.2).

Three buffers, TBS, PBS and carbonate, with and without 0.05% Tween 20, were evaluated for their effects on the sensitivity of DIBA, DAS-ELISA and DAS-indirect ELISA with two antibodies and three CTV isolates. In DIBA, PBS gave the








84

strongest reactions, followed by TBS and carbonate. The addition of Tween 20 gave slightly weaker reactions with all buffers, but the spots were more uniformly spread on the nitrocellulose. With both ELISA procedures the presence of Tween 20 in the extraction buffers tended to increase the OD405 values slightly. However, in DAS-indirect ELISA, the presence of Tween 20 in the PBST more than doubled OD405 readings over PBS alone for CTV isolates T62a and T66a, but had little effect with isolate T26. This did not occur with TBS or the carbonate buffer. One possible explanation is that the presence of Tween 20 in the PBST caused a conformational change in the coat protein of T62a and T66a exposing more of the epitope for binding with the 3DF1 MCA. This appeared to occur to a much lower degree with isolate T26 in DAS-indirect ELISA and with all three of these isolates reacting with the polyclonal antibodies in DAS-ELISA. These results indicate that PBS might be the best extraction buffer for DIBA, and PBST or TBST might be the best for ELISA. However, it would always be advisable to determine the effects of Tween 20 on the serological methods and antigen/antibody combinations under investigation.

There are several advantages to using DIBA over conventional DAS-ELISA or DAS-indirect ELISA for CTV detection. DIBA was rapid and easy to perform and, it was as sensitive as either ELISA procedure for CTV diagnosis. The entire test could be performed in 2-3 hours using polyclonal








85

antibodies (slightly longer with monoclonal antibodies), and minimal laboratory equipment was needed. The use of small containers such as the polypropylene covers for incubation and washing steps enabled the recovery and re-use of both virus specific antibodies and goat anti-species IgG conjugates for at least four weeks when the solutions were properly preserved with 0.02% sodium azide and stored at SoC between uses. The template constructed from the rack of a micropipet tip holder box enabled the uniform spacing of samples on the nitrocellulose membranes.

As pointed out by Powell (1987), one disadvantage of DIBA as compared to ELISA procedures is the lack of quantitative measurements. However, with appropriate positive and negative controls, DIBA can be used reliably for routine diagnostic work where no quantitative measurements are needed. When quantitation of CTV antigens is required, ELISA should be used.














CHAPTER 5
SUMMARY AND CONCLUSIONS



Four different naturally occurring Florida mild citrus tristeza virus (CTV) isolates were evaluated under greenhouse conditions at two temperature regimes for their crossprotecting ability against the induced decline by a CTV severe challenge isolate in two susceptible scion/rootstock combinations. DAS-ELISA with polyclonal antisera was used to determine the total antigen titer in plants inoculated with mild isolates and those challenged with the severe isolate. The MCA-13 strain specific monoclonal antibody was successfully used in DAS-indirect ELISA for differential isolate detection and quantitation of the severe challenge isolate in mixed infections.

The effectiveness of the graft transmission of CTV was evaluated with three CTV isolates by using leaf and bark tissue from three citrus donor hosts to three receptor hosts. The distribution of the virus in different plant tissues was also studied.

A dot-immunobinding assay (DIBA) was adapted for CTV detection. The sensitivity level of DIBA was evaluated using three different polyclonal and two monoclonal antibodies and



86








87

compared with sensitivities of double antibody sandwich (DAS) ELISA and DAS-indirect ELISA with 12 different CTV isolates. The conclusions of this research are summarized as follows: Cross-Protection of Citrus Tristeza Virus

At warm temperatures plants pre-inoculated with mild isolates showed relatively low optical density (OD405) values in the range of 0.091-0.145 in DAS-ELISA with polyclonal antisera; whereas, at cool temperatures the values were commonly in the range of 0.300-0.400. At warm temperatures, the MCA-13 monoclonal antibody in plants pre-inoculated with mild isolates but unchallenged, gave OD405 values in the range of 0.030-0.045 which were as low as the uninoculated control plants. However, at cool temperatures, the OD405 values obtained with mild isolates were slightly higher than those obtained at warm temperatures, indicating that the MCA-13 monoclonal antibody may react to some extent with mild isolates when they are above a certain titer in the plants. Nevertheless, the OD405 values were always lower than 0.100, which was considered a negative reaction.

When plants pre-inoculated with mild isolates and further challenged with the severe isolate were assayed with the MCA13 monoclonal antibody, they gave typically lower OD405 values for the T66a isolate than the unprotected challenged control plants which did not have mild isolates. In this regard, the T26 and T30 isolates gave the more uniform lower OD405 values for the T66a isolate in all evaluated treatments. This would




Full Text
70
1/10
1/20
1/4 0
1/80
1/160
1/320
1/640
TBS-Tween
Fig. 4.3 Relative sensitivity level of different polyclonal
and monoclonal antibodies specific to citrus tristeza virus
(CTV) in dot-immunobinding assay. Rows A, B and C, polyclonal
antibodies nos. 1051, 1052, and 1053, respectively. Rows D
and E, monoclonal antibodies 3DF1 and MCA-13, respectively.
Extract of Citrus excelsa greenhouse grown plants infected
with CTV T36 isolate was prepared in TBS-Tween 1:10 (w/v) and
two-fold diluted with extract of healthy plants. Reaction
conditions are described in materials and methods.


2
23,000 and 21,000 (Lee et al. 1988b), respectively, have been
associated with CTV virions. The virus is readily transmitted
by budding and grafting (Bennett and Costa, 1949; Bar-Joseph
and Lee, 1990; Bar-Joseph et al. 1979a). Mechanical
transmission has been accomplished by slash inoculation of
partially purified virus preparations into the stem of hosts
such as citron (Citrus medica L.) and Mexican lime {C.
aurantifolia (Christm.) Swing.} (Garnsey and Muller, 1988;
Garnsey et al. 1977; Muller and Garnsey, 1984). Seed
transmission has not been demonstrated (McClean, 1957;
Wallace, 1978).
Citrus tristeza virus occurs naturally with a diversity
of isolates or strains which may differ greatly in their
biological properties, such as symptomatology in different
citrus hosts (Garnsey et al. 1987; McClean, 1974), aphid
transmissibility (Bar-Joseph and Loebenstein, 1973; Bar-Joseph
et al. 1977; Roistacher, 1981; Yokomi and Garnsey, 1987), and
sensitivity to warm temperatures (Ieki and Yamada, 1980;
Roistacher et al. 1974).
Several sensitive and relatively rapid methods have been
developed to diagnose the presence of CTV in infected plants.
These methods include SDS-immunodiffusion procedures (Garnsey
et al. 1979; Bar-Joseph et al. 1980), enzyme-linked
immunosorbent assay (ELISA) (Bar-Joseph et al. 1979b, 1980),
light and electron microscopy (Brlansky, 1987; Brlansky et al.
1984; Garnsey et al. 1980a), and in situ immunofluorescence


47
differences (P < 0.05%) for all donor-receptor host
combinations. Likewise, according to Duncan's multiple range
comparison test, there were statistical differences between
some of the hosts tested (Table 3.2).
The rate of transmission for the three different CTV
isolates tested with each donor host is shown in Table 3.3.
The T26 isolate was transmitted at a rate of 69.0% to 92.8%.
The transmission rates of T30 and T66a isolates ranged from
62.5% to 100% and from 71.4% to 96.7%, respectively, from the
hosts tested. While there were statistical differences in
the rates of transmission for some of the virus isolate/donor
host combination, the overall average of transmission showed
no significant differences among them (Table 3.3).
The overall statistical analysis for percent transmission
of the interactions among the different donor/receptor/virus
isolate/inoculum pieces combinations, indicated no significant
differences for receptor and virus isolates alone, and for the
combinations of donor/inoculum pieces, receptor/inoculum
pieces, and for donor/receptor/virus isolate. However,
significant differences (P < 0.05) were found for donor and
inoculum pieces alone, and for the interactions between
donor/receptor, donor/virus isolate, receptor/virus isolate,
and virus isolate/inoculum pieces.
Virus distribution and antigen concentration in host
tissues. The relative antigen titer of CTV as measured by
DAS-ELISA in each host tissue/virus isolate combination is


36
Likewise, a mixture of several severe isolates should be used
as the challenge virus source to give a better evaluation of
decline symptoms under greenhouse conditions.
The methodology described herein offers the following
advantages: i) Depending upon space availability, large
numbers of mild isolates can be evaluated uniformly in a time
period of 18 to 24 months: from six to twelve months to get
the one-year-old plants infected with the mild isolates and
verification of infection by serology, two months for
challenge and ten months for final evaluation; ii) The
availability of the MCA-13 monoclonal antibody provides a
useful tool to detect the presence of a severe isolate in the
challenged plants, and at the same time allows an estimate of
the relative ability of mild isolates to prevent the
establishment of the severe isolate in the challenged plants;
iii) Mild isolates can be evaluated in grenhouses without the
risks that represent the threat of recurrent freezes
especially in Florida in recent years, the lack of an
appropriate natural challenge pressure (vector or severe
isolate), and the effect of some other devastating diseases
(i.e. greening or blight) that can hamper the reliable
evaluation of cross-protection experiments under field
conditions. Some limitations in the methodology can be also
visualized. The use of leaf piece grafts is not always highly
efficient to transmit CTV from the donor propagation hosts to
the receptor test plants (see Chapter 3) This leaves the



55
at higher concentrations in young phloem-rich tissues
(Garnsey, et al. 1979; Bar-Joseph, et al. 1979a); however,
the serological titer frequently decreases as the tissues
reach maturity or when the plants are exposed to warm
environments (Garnsey et al. 1981a; Lee et al. 1988c). The
OD405 values obtained in this research were low if compared
with those found when DAS-ELISA is used routinely for CTV
diagnosis (Bar-Joseph et al. 1979b; Garnsey et al. 1980b);
however, this part of the work was addressed to determine the
virus titer in the tissues suitable for graft transmission,
and young tender tissue sometimes is not a good source of
inoculum for leaf piece grafts (personal observations).
The overall analysis of the results obtained indicates
that the efficiency of the graft transmission of CTV is
conditioned primarily by the donor/receptor host combination,
and secondly by the virus isolate involved, but apparently not
by the interaction of the three. For example, C. excelsa
showed an overall rate of transmission in the range of 72.5%
with all receptor hosts tested (Table 3.2), and a similar low
pattern between 69% and 76.7% (= 72.8%) was obtained for the
three isolates tested (Table 3.3). Likewise, when Madam
Vinous was used as the donor host, there was an overall rate
of transmission of 90.6% (Table 3.2). A rate of 77.3-100% (=
88.6%) occurred from this host with the three isolates tested
(Table 3) A comparable event was also scored when Mexican
lime was the donor host (Table 3.2 and 3.3). The statistical


32
only with mild isolates (Table 2.5 and 2.6). In regard to
Valencia/sour orange plants pre-inoculated with mild isolates
and challenged with the T66a severe isolate, the lowest
decline index (2.6) and lowest number of dead plants (0/5)
was obtained with the T26 isolate. Whereas, the highest
decline index (6.3) and highest number of dead plants (3/8)
was obtained with the Tila isolate. A decline index of 4.0
and 1/6 dead plants were scored with the uninoculated and
challenged control plants (Table 2.5). Also at warm
temperatures, Valencia/macrophylla plants pre-inoculated with
mild isolates and challenged with the T66a severe isolate, the
T26 and T55a isolates obtained the lowest decline index (4.2
and 2.2) and lowest number of dead plants (0/5 and 0/2),
respectively.
At cooler temperatures, there were some differences in
the effect of the CTV mild isolates on growth of both
Valencia/sour orange and Valencia/macrophylla. The decline
indexes for the healthy uninoculated control plants were
variable and ranged from values below to above those obtained
with plants inoculated only with mild isolates (Table 2.7 and
2.8). In general, there was a remarkable growth reduction
effect at cool temperatures even on healthy uninoculated
control plants. In this regard Valencia/sour orange plants
pre-inoculated with mild isolates and challenged with the T66a
severe isolate, the T26 and T55a isolates obtained the lowest
decline index (5.5 and 5.0) and lowest number of dead plants


69
and stronger reactivity for positive samples. Either the
absence of PVP or BSA concentrations of less than 1% in the
antibody diluent frequently resulted in a dark background on
the nitrocellulose when 3% BSA was used for blocking.
Evaluation.- The relative sensitivity level of DIBA was
measured using increasing dilutions of a plant extract
containing CTV isolate T36. This is shown in Figure 4.3.
Polyclonal IgGs Nos. 1051, 1052, and 1053 (1.0 xg/m 1 IgG for
30 min) consistently gave a distinct positive reaction with
CTV T36 diluted to 1/160. There was a slight but consistently
positive reaction at the 1/320 dilution and occasionally at
1/640 (Fig. 4.3 A, B, C) The reactivity of both 3DF1 and
MCA-13 MCAs (1.0 /g/ml and 1:5,000, respectively) was also in
the range of 1/160 and 1/320, but only when the incubation
time was extended to 18 hr (Fig. 4.3 D, E).
The relative sensitivities of DAS-ELISA and DAS-indirect
ELISA were measured similarly against the diluted extract of
CTV T36. These results are summarized in Table 4.1. DAS-
ELISA, using polyclonal IgG No. 1053 for both coating and
conjugate steps, gave OD405 value of 0.109 (i.e above 0.100)
with the CTV T36 sample diluted to 1/160. This was considered
a positive reaction. Thus the OD405 value of 0.070 at a 1/320
dilution was considered a negative reaction. DAS-indirect
ELISA, using polyclonal IgG No. 1053 for coating and 3DF1 MCA
as second antibody in the double sandwich, gave a positive and
negative reaction at dilutions of 1/320 and 1/640,


Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
CITRUS TRISTEZA VIRUS: CROSS-PROTECTION, GRAFT
TRANSMISSION, AND SEROLOGY
By
MARIO ALBERTO ROCHA-PEA
DECEMBER 1990
Chairperson: Dr. Richard F. Lee
Major Department: Plant Pathology
The objectives of this research were i) to evaluate some
citrus tristeza virus (CTV) mild isolates under greenhouse
conditions for cross protecting ability against the decline
syndrome, and ii) to develop methods for detection of severe
CTV challenge isolates in mixed infections in cross-protection
experiments. Valencia sweet orange plants budded on sour
orange rootstock were graft-inoculated by leaf pieces using
any of four different mild CTV isolates and subsequently
graft-challenged with a severe CTV isolate. Treatments were
evaluated at temperature regimens of 21-38C and 21-33C.
Plants pre-inoculated with mild isolates when challenged with
the severe isolate gave relatively lower ELISA values as
compared to the unprotected, challenged control plants. The
MCA-13 monoclonal antibody provided a rapid method to detect
xi


Table 4.2 Comparison of DIBA, DAS-ELISA and DAS-indirect ELISA for the detection of
citrus tristeza virus (CTV) and relative reactivity of polyclonal (1053) and monoclonal
(3DF1 and MCA-13) antibodies with CTV isolates.
Virus
DIBA
isolate'
1053
3DF1
MCA-13
Tila
(M)1/
+1/
+
-
T2 6
(M)
+
+/-
-
T30
(M)
+
+
-
T50a
(M)
+
+
+/-
T55a
(M)
+
+
+/-
T3
(S)
+
+
+
T4
(HD)
+
+
+/-
T36
(S)
+
+
+
T62a
(S)
+
+
+/-
T65a
(S)
+
+
+
T66a
(S)
+
+/-
+
T67a
(S)
+
+
+
DAS-ELISA1^ DAS-indirect ELISA1/
1053
3DF1
MCA-13
0.204
+5/
1.568
+
0.091
0.137
+
0.053
-
0.057
0.227
+
1.936
+
0.090
0.498
+
2.584
+
0.120
0.443
+
2.018
+
0.119
0.404
+
2.468
+
0.522
0.304
+
1.992
+
0.092
0.572
+
2.628
+
0.892
0.556
+
2.045
+
0.115
0.556
+
2.527
+
0.745
0.588
+
0.191
+
0.735
0.250
+
2.235
+
0.480
ui


88
indicate that the mild isolates, especially the T26, T30, and
T55a, prevented to some extent the multiplication of the
challenge severe isolate, and that they were apparently
working as cross-protecting agents. The evaluation of the
virus titer of the T66a severe isolate with the MCA-13
monoclonal antibody in the treatments pre-inoculated with mild
isolates and further challenged provided a measurable
parameter to estimate the ability of mild isolates to prevent
the establishment the severe isolate.
The effect of mild CTV isolates on performance of both
Valencia/sour orange and Valencia/macrophylla was variable.
At warm temperatures, the effect was negligible. The decline
index for the healthy uninoculated control plants was nearly
equal to or greater than those inoculated only with mild
isolates. However, at cool temperatures there was a
remarkable growth reduction effect in all treatments
evaluated. The decline indexes for the healthy uninoculated
control plants was variable and ranged from values below to
above those obtained with plants inoculated only with mild
isolates. Of all treatments evaluated at both temperatures,
the T26 isolate, and also T55a to some degree, obtained the
lowest decline index scores and lowest number of dead plants
as compared with the uninoculated challenged control plants.
This provided further evidence of their cross-protecting
effect, especially the T26 isolate against the development of
the CTV-ID syndrome.


40
numbers of plants are to be inoculated with limited sources
of inoculum (Garnsey and Whidden, 1970).
During several experiments with CTV in Florida (Chapter
2; Rocha-Pea et al. 1990) large numbers of plants were
inoculated by leaf piece grafts with several CTV isolates that
were propagated in different citrus hosts. There were notable
differences in the efficiency of transmission of some CTV
isolates from different donor hosts, and in some cases no
transmission was achieved even after repeated inoculations.
The objectives of this research were to evaluate the effect
of different citrus hosts on the efficiency of graft
transmission of CTV, and to determine the relative
distribution of the virus in different host tissues.
Materials and Methods
Virus isolates and donor hosts. Three isolates of CTV,
T26, T30, and T66a, were used throughout the study. They have
been described previously (Garnsey et al. 1987; Lee, 1984;
Yokomi and Garnsey, 1987). Virus isolates were propagated in
Citrus excelsa Wester, Mexican lime {C. aurantifolia
(Christm.) Swingle} and Madam Vinous sweet orange {C. sinensis
(L.) Osb.} plants, herein referred to as donor hosts,
maintained in a greenhouse with mean minimum and maximum
temperatures of 21 and 33C, respectively. Inoculum tissue
from donor hosts was evaluated by serological indexing by the
double antibody sandwich enzyme-linked immunosorbent assay
(DAS-ELISA) (see below) to verify the presence of CTV before


21
the Valencia/macrophylla combination (Table 2.4) because not
enough plants were available.
From the standard curve prepared with purified T26 (Fig.
2.1), it was estimated that an OD405 value of 0.638 was
approximately equivalent to 20 g/ml of CTV antigen, assuming
an extinction coefficient of 2.0 (Gonsalves et al. 1978).
Therefore, there was an average of 3.1 ng of CTV antigen per
every 100 mg of tissue for each 0.100 OD405 value in the test
samples.
Detrimental effects of mild isolates and evaluation of
cross protection. The protecting effect of mild isolates was
evaluated on the basis of their ability to prevent the
detrimental effects on stem diameter, plant growth and foliage
symptoms caused under greenhouse conditions by the T66a severe
decline isolate. The detrimental effects of mild isolates
alone also were evaluated. The number of plants and the
scores for the decline index established in every treatment
are shown in Tables 2.5, 2.6, 2.7, and 2.8. At warm
temperatures, Valencia/sour orange plants inoculated with mild
isolates and unchallenged with T66a, and the healthy
uninoculated controls, gave overall decline index values in
the range of 0.0 and 1.5. Whereas, the plants pre-inoculated
with mild isolates and challenged with T66a showed higher
decline index values of 6.3, 2.6, 4.2, and 5.5 for the Tila,
T26, T30, and T55a mild isolates, respectively. The control
plants uninoculated with mild isolates but challenged with


39
CTV from Meyer lemon to 25 of 27 Mexican lime and sour orange
seedlings when leaf pieces contained midribs. However, the
efficiency of transmission decreased more than 50% when grafts
did not include the leaf midrib. A modification of Cohen's
procedure, called "leaf-disc grafting" (Blue et al. 1976)
involved the use of circular leaf pieces 6 mm in diameter cut
from the midrib area of a donor plant leaf and placed into a
corresponding hole in the receptor plant leaf. The midrib of
the donor tissue is aligned with that of the receptor leaf,
and grafts are held in place with transparent tape. This
method was as successful as bud inoculation for transmitting
many CTV isolates to Mexican lime plants from different citrus
species, and was more efficient for transmitting mild CTV
isolates. The leaf-disc method was used for routine indexing
in the citrus budwood certification program in California
(Calavan et al. 1978).
Another method involving the use of leaf piece grafts
was reported by Garnsey and Whidden (1970). Rectangular leaf
pieces from infected plants were inserted under corresponding
s .
rectangular bark flaps cut in the stem of receptor hosts.
This procedure has been used widely for many years with CTV
and other citrus viruses, and it has been used in the
characterization of the biological properties of diverse
worldwide collection of CTV isolates (Garnsey et al. 1987).
Leaf piece grafts are especially advantageous when large


85
antibodies (slightly longer with monoclonal antibodies), and
minimal laboratory equipment was needed. The use of small
containers such as the polypropylene covers for incubation and
washing steps enabled the recovery and re-use of both virus
specific antibodies and goat anti-species IgG conjugates for
at least four weeks when the solutions were properly preserved
with 0.02% sodium azide and stored at 5C between uses. The
template constructed from the rack of a micropipet tip holder
box enabled the uniform spacing of samples on the
nitrocellulose membranes.
As pointed out by Powell (1987) one disadvantage of DIBA
as compared to ELISA procedures is the lack of quantitative
measurements. However, with appropriate positive and negative
controls, DIBA can be used reliably for routine diagnostic
work where no quantitative measurements are needed. When
quantitation of CTV antigens is required, ELISA should be
used.


81
of BSA of less than 1% or the absence of PVP in the antibody
diluent frequently resulted in the nitrocellulose membrane
turning dark when BSA was used for blocking. Thus TBS
containing 1% BSA and 2% PVP was always used as the antibody
diluent.
Differences were found in the reactivities of the
antibodies tested. All polyclonal IgGs (Nos. 1051, 1052, and
1053) reacted similarly in DIBA. After 30 min incubation (at
1.0 jul/ml IgG) they gave strong positive reactions with CTV-
infected samples, but none with healthy samples. However,
incubations of 60 min or more sometimes resulted in the
occurrence of nonspecific reactions with extracts of healthy
plants. This was prevented when the polyclonal antibodies
were cross-absorbed with extracts of healthy plants before
use. These antibodies gave a strong positive reaction with
the 12 selected CTV isolates tested, and had a sensitivity
(highest dilution which gave positive reaction) limit of 1/320
with CTV T36 infected samples.
The 3DF1 MCA when used at an IgG concentration of 1.0
/xg/ml reacted moderately with most of the CTV isolates tested
and had a relative sensitivity limit of 1/320 in DIBA and
1/680 in DAS-indirect ELISA. However, in some cases it was
necessary to extend the incubation time (15-18 hr) with 3DF1
in DIBA, because it had a reactivity level that was somewhat
lower than any of the polyclonal antibodies tested.


101
Permar, T.A., Garnsey, S.M., Gumpf, D.J., and Lee, R.F. 1990.
A monoclonal antibody that discriminates strains of
citrus tristeza virus. Phytopathology 80:224-228.
Powell, C.A. 1987. Detection of three plant viruses by dot-
immunobinding assay. Phytopathology 77:306-309.
Purcifull, D.E. and Batchelor, D.L. 1977. Immunodiffusion
tests with sodium dodecyl sulfate (SDS)-treated plant
virus and plant inclusions. Technical Bulletin 788.
Florida Agricultural Experiment Station. 39 p.
Rocha-Pea, M.A. and Lee, R.F. 1991. Serological techniques
for detection of citrus tristeza virus. J. Virol. Meth.
(in press).
Rocha-Pea, M.A., Lee, R.F., Permar, T.A., Yokomi, R.K., and
Garnsey, S.M. 1990. Use of enzyme-linked immunosorbent
and dot-immunobinding assays to evaluate two cross
protection experiments after challenge with a severe
citrus tristeza virus isolate. In: Proc. 11th Conf.
Intern. Organ. Citrus Virol. Riverside, California (in
press).
Roistacher, C.N. 1981. A blueprint for disaster: Part II.
Changes in the transmissibility of seedling yellows.
Calif. Citrograph 67:29-32.
Roistacher, C.N., Blue, R.L., Nauer, E.M., and Calavan, E.C.
1974. Suppression of tristeza virus symptoms in Mexican
lime seedlings grown at warm temperatures. Plant Dis.
Rep. 58:757-760.
Roistacher, C.N., Dodds, J.A., and Bash, J.A. 1987. Means of
obtaining and testing protective strains of seedling
yellows and stem pitting tristeza virus: A preliminary
report. Phytophylactica 19:199-203.
Roistacher, C.N., Dodds, J.A., and Bash, J.A. 1988. Cross
protection against citrus tristeza seedling yellows and
stem pitting viruses by protective isolates developed in
greenhouse plants. Pages 91-100, in: Timmer, L.W.,
Garnsey, S.M., and Navarro, L. (eds). Proc. 10th Conf.
Intern. Organ. Citrus Virol. Riverside, California.
Schwartz, R.E. 1968. Transmission of the tristeza virus by a
leaf union method. S. African J. Agr. Sci. 11:617-622.
Spinola, S.M. and Cannon, J.G. 1985. Different blocking agents
cause variation in the immunologic detection of proteins
transferred to nitrocellulose membranes. J. Virol. Meth.
81: 161-165.


xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID E4W1QATXR_O8QPBB INGEST_TIME 2015-02-17T19:41:05Z PACKAGE AA00028765_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES


14
of healthy C. excelsa was used to estimate the relative
antigen concentration of test samples.
Detrimental effects of mild isolates and evaluation of
cross protection. To evaluate the effect of each CTV isolate
on the inoculated plants and the protecting ability of the
mild isolates against the T66a challenge isolate, evaluations
were made at five and ten months after the challenge
inoculation with the T66a isolate. Phloem necrosis was
evaluated by cutting a bark flap at the bud union and the
plant tissue was examined with a hand lens for browning. A
decline index was assigned for each plant. The parameters
scored were stem diameter, plant growth, and foliage symptoms
for decline. Each parameter was visually rated from 0
(minimum) to 3 (maximum), for a maximum cumulative score of
9 for each plant. A high decline index sometimes was
accompanied by plant death. The decline index for each
treatment was the average of the cumulative scores for all
plants in that treatment.
Results
Antigen titers of mild and severe CTV isolates with
polyclonal and MCA-13 monoclonal antibodies. The antigen
titers expressed as optical density (OD405) values for the
different temperature and host treatments, measured by DAS-
ELISA with polyclonal antibodies (PCA) and DAS-indirect ELISA
with MCA-13, are summarized in Tables 2.1, 2.2, 2.3, and 2.4.
At warm temperatures, Valencia/sour orange plants inoculated


63
405 nm (OD405) using a Labinstruments model EAR 400 AT ELISA
plate spectrophotometer.
DAS-indirect ELISA.- For DAS-Indirect ELISA the plates
were coated with polyclonal IgG No. 1053, and antigen samples
were added as described for DAS-ELISA. The 3DF1 MCA was used
at a concentration of 0.2 /g/ml IgG and MCA-13 ascites fluid
was used at a dilution of 1:5,000 (v/v). Each was diluted in
conjugate buffer, added to the appropriate plates and
incubated for 4 hr 37C. After washing, alkaline phosphatase
conjugated goat anti-mouse IgG was added at a dilution of
1:7,500 (v/v) in conjugate buffer and incubated for 2 hr at
37C. The enzyme-substrate reaction was carried out as
described for DAS-ELISA.
Evaluation.- Twelve selected naturally-occurring Florida
CTV isolates with different biological properties (Garnsey et
al. 1987; Permar et al. 1990; Rocha-Pea and Lee, unpublished)
were used to determine the range of reactivity of each
antibody. All virus source plants were maintained in a
greenhouse with mean minimum and maximum temperatures of 21
and 38C, respectively. The isolates Tila, T26, T30, T50a and
T55a produce very mild symptoms and little or no stunting in
Mexican lime seedlings {Citrus aurantifolia (Christm.)
Swingle}, no seedling yellows on sour orange (C. aurantium L.)
or grapefruit (C. paradisi Macf.) and no symptoms in sweet
orange {C. sinensis (L.) Osb.} and sweet/sour orange
combinations. The isolate T4 causes a moderate or stronger


77
Evaluation of different buffers for sample extraction.-
The reactions of the polyclonal IgG 1053 and MCA 3DF1 with
the different CTV isolate/extraction buffer combinations in
DIBA are shown in Figure 4.5. In general, there was a strong
positive reaction with all the CTV isolate/extraction buffer
combinations tested, with the strongest reaction occurring
when PBS without Tween 20 was used as the extraction buffer
(Fig. 4.5). The presence of Tween 20 in the extraction buffer
gave a weaker but relatively more uniform reaction spot with
both antibodies.
The results of the evaluation of the different extraction
buffers by DAS-ELISA and DAS-indirect ELISA are shown in
Figure 4.6. In DAS-ELISA the strongest reactions occurred
with isolate T62a, followed by T26 and T66a, respectively.
The presence or absence of Tween 20 made little difference in
DAS-ELISA (Fig. 4.6 left). In DAS-indirect ELISA with the
3DF1 MCA the strongest reactions occurred with isolate T26,
followed by T62a and T66a, respectively. The presence of
Tween 20 gave slightly stronger reactions in most cases. With
PBS, the presence of Tween 2 0 more than doubled the OD405
values for isolates T62a and T66a, but caused only a slight
increase for isolate T26 (Fig. 4.6 right)
Discussion
The dot-immunobinding assay was adapted for detection of
CTV. This included the testing of different agents for
blocking and as diluents for the CTV specific antibodies and


extracts of greenhouse-grown Madam Vinous
sweet orange plants infected with the CTV
isolates ground at a 1:10 dilution 78
Figure 4.6 Evaluation of different extraction
buffers on the sensitivity of DAS-ELISA
and DAS-indirect ELISA with citrus tristeza
virus (CTV) isolates T26, T62a and T66a.
TBS = Tris buffered saline, TBST = TBS+Tween
(0.05%), PBS = phosphate buffered saline,
PBST = PBS+Tween (0.05%), Carb = carbonate
buffer, CarbT = Carb+Tween (0.05%),
HC = healthy control. The IgG of
polyclonal antibodies No. 1053 was used
to coat the plates for both DAS-ELISA and
DAS-indirect ELISA. For DAS-ELISA the No.
1053 IgG conjugate was used as second
antibody. For DAS-indirect ELISA the
unlabeled 3DF1 monoclonal antibody was the
intermediate antibody followed by the goat
anti-mouse IgG conjugate. Samples were
200 nl of buffer extracts of
greenhouse-grown Madam Vinous sweet orange
plants infected with the CTV isolates
ground at a 1:10 dilution 79
viii


53
93.1.7% transmission to grapefruit and Madam Vinous,
respectively and a 76.9% rate to Mexican lime. Transmission
from Madam Vinous sweet orange was between 84.6 and 100% in
all receptors tested. It was surprising that transmission
rates between the same species were only 86.7% for Madam
Vinous, and 76.9% for Mexican lime (Table 3.2).
Madam Vinous sweet orange was the most efficient donor
host with the three receptor hosts tested (90.6%), followed
by Mexican lime (85.2%). C. excelsa was a poor donor host
(72.5%), being relatively efficient only when inoculated to
Mexican lime (Table 3.2).
Previous studies on the transmission of CTV by grafting
procedures have shown that a period of at least ten days
contact between grafted tissues is needed to obtain
transmission of the virus to the receptor host (Tolba et al.
1976; Yamaguchi and
Patpong,
1980). In
this
study,
the
survival of
grafted
tissue
was scored
21
days after
inoculation,
but the
inoculated tissue
was left in
the
receptor plants for up to five months. This should have been
ample time for contact between the inoculum and receptor
cambium to establish a tissue union with a subsequent
transmission of CTV. Furthermore, when leaf pieces were used
as inoculum, a small portion of the midrib was included in
every piece to increase the success of the grafting. The
overall rates of successful grafts were about 83% and a 66.7%
for leaf and bark pieces, respectively (Table 3.1).


98
Gonsalves, D. and Garnsey, S.M. 1989. Cross-protection
techniques for control of plant virus diseases in the
tropics. Plant Disease 73:592-597.
Gonsalves, D., D.E. Purcifull, and Garnsey, S.M. 1978.
Purification and serology of citrus tristeza virus.
Phytopathology 68:553-559.
Grant, T.J., Moreira, S., and Salybe, A.A. 1961. Citrus
variety reaction to tristeza virus in Brazil when used
in various rootstock and scion combinations. Plant Dis.
Rep. 45:416-421.
Guerri, J., Moreno, P., and Lee, R.F. 1990. Identification of
citrus tristeza virus strains by peptide maps of virion
coat protein. Phytopathology 80:692-698.
Gumpf, D.J., Zheng, G.-Y., Moreno, P., and Diaz, J.M. 1987.
Production and evaluation of specific monoclonal
antibodies to citrus tristeza virus strains.
Phytophylactica 19:159-161.
Hamilton, R.I. 1985. Using plant viruses for disease control.
HortScience 20:848-852.
Hawkes, R. Niday, E. and Gordon, J. 1982. A dot-
immunobinding assay for monoclonal and other antibodies.
Anal. Bioch. 119:142-147.
Hibi, T. and Saito, Y. 1985. A dot-immunobinding assay for the
detection of tobacco mosaic virus in infected tissues.
J. Gen. Virol. 66:1191-1994.
Ieki, H. 1989. The use of cross-protection with mild strains
of citrus tristeza virus (CTV) to control stem pitting
disease of citrus in Japan. Pages 8-14. FFTC Extension
Bulletin No. 284.
Ieki, H. and Yamada, S. 1980. Inactivation of citrus tristeza
virus (CTV) with heat treatment: Heat tolerance and
inactivation of CTV on rootstock-scion combinations.
Pages 220-224, in: Calavan, E.C., Garnsey, and Timmer,
L.W. (eds). Proc. 8th Conf. Intern. Organ. Citrus Virol.
Riverside, California.
Irey, M.S., Permar, T.A., and Garnsey, S.M. 1988.
Identification of severe isolates of citrus tristeza
virus in young field plantings by enzyme-linked
immunosorbent assay. Proc. Fla. State Hort. Soc. 101:73-
76.


67
slight pink reaction was consistently obtained with sap from
healthy plants (Fig. 4.1 E).
Polyclonal IgGs had to be cross absorbed with buffer
extracts of healthy tissue to reliably discriminate between
healthy and CTV infected samples (Fig. 4.2 A, B) and
reactivity varied with concentration and incubation times.
The strongest reactions and best discrimination between CTV
positive and negative samples were achieved using 1.0 ng
IgG/ml and 30 min incubation. An incubation time of 60 min
with 1.0 or 0.1 ng IgG/ml, or longer (18 hr) even with as low
as 0.01 jug IgG/ml frequently resulted in the occurrence of
nonspecific reactions with healthy samples, even when the IgG
had been cross absorbed with extracts of healthy plants. sA
would be expected, MCAs did not have to be cross absorbed with
buffer extracts from healthy plants prior to use. The
reactivity of the 3DF1 and MCA-13 MCAs was substantially lower
than that obtained with any of the polyclonal IgGs tested.
For both 3DF1 (1.0 /xg/ml) and MCA-13 (diluted 1:5,000), the
incubation time was extended to 2 hr or longer (18 hr) to
achieve an adequate positive reaction with infected samples
(Fig. 4.2 C, D). The reactivity of 3DF1 was always slightly
lower than that obtained with either polyclonal IgG or the
MCA-13.
In initial experiments with polyclonal IgGs, 1% BSA in
TBS plus 2% PVP was chosen as the antibody diluent because it
consistently gave a whiter background on the nitrocellulose


2/
Buffer extract (1:10 dilution) of bark from greenhouse grown C. excelsa plants
infected with CTV T36 successively two-fold diluted with healthy C. excelsa buffer
extract.
3 /
' Visual evaluation for presence of a purple color (+ = positive, +/- = inconclusive, -
= negative.
Optical density at 405 nm (OD4Q5) after 60 min of substrate reaction. Mean of two
replicatipns per plate. Reactions were considered positive (+) when OD4Q5 values were
higher than three times the mean of healthy controls or 0.100, whichever was greater.
Reactions with lower values were considered negative (-).


56
significance found for the interactions donor/receptor and
donor/virus isolate, and no significance for the interaction
of donor/receptor/virus isolate supports this conclusion.
The use of leaf and/or bark pieces for graft transmission
of CTV may be advantageous when large numbers of plants are
to be inoculated with limited sources of inoculum (Garnsey and
Whidden, 1970). However, in the light of the results of this
research, in order to achieve a high level of transmission,
the efficiency of the donor host and the donor/receptor host
combination should be considered.


CHAPTER 3
EFFECTIVENESS OF CITRUS SPECIES AS DONOR HOSTS FOR
GRAFT TRANSMISSION OF CITRUS TRISTEZA VIRUS
Introduction
Citrus tristeza virus (CTV) has long been known to be
transmitted by budding and by different grafting procedures
(Bennett and Costa, 1949; Bar-Joseph et al. 1979a; Bar-Joseph
and Lee, 1990). In 1951, Wallace experimentally transmitted
CTV by placing small portions of donor leaf or bark tissue
under a flap of bark on receptor plants. By this method, CTV
was transmitted from many field sources of sweet orange to
Mexican lime receptor plants, and from Mexican lime to healthy
sweet orange plants (Wallace, 1951). Schwartz (1968)
transmitted CTV by connecting the distal portion of the leaf
of infected plants to a matching proximal part on a leaf of
a receptor plant. By this method, CTV was transmitted to 9
of 20 Mexican lime plants, but transmission was obtained only
when callus formation occurred between grafted tissues; also,
older and dark green leaves were a better source than younger
leaves for both callus formation and virus transmission.
Cohen (1972) described a method for CTV transmission in
citrus by grafting triangular leaf pieces into triangular
holes cut in the leaves of receptor plants. He transmitted
38


33
(0/2 and 1/4), respectively. Whereas, the highest decline
index (7.5) and highest number of dead plants (3/4) were
obtained with the Tila isolate. A decline index of 7.5 and
1/2 dead plants were scored with the uninoculated and
challenged control plants (Table 2.7). At cool temperatures,
Valencia/macrophylla plants pre-inoculated with mild isolates
and challenged with the T66a severe isolate, had decline index
scores similar to the plants inoculated only with mild
isolates. No dead plants were obtained in this portion of the
experiment (Table 2.8).
Of all treatments evaluated at both temperatures, the
T26 isolate, and also T55a to some degree, obtained the lowest
decline index scores and lowest number of dead plants as
compared with the uninoculated challenged control plants.
This provided further evidence of their cross-protecting
effect, especially the T26 isolate against the development of
the CTV-ID syndrome.
Of special interest was the high decline index scores
and number of dead plants in plants pre-inoculated with the
Tila mild isolate and further challenged with the T66a
isolate. It seemed that the combination of Tila and the T66a
isolates produced a more severe reaction on the challenged
plants than that caused by the T66a isolate alone in the
unprotected control plants. The lack of cross-protecting
ability of the Tila isolate has been previously reported
(Yokomi et al. 1987).


Table 2.6 Effect of citrus tristeza virus (CTV) mild isolates on the development of
the CTV decline syndrome in plants unchallenged and challenged by the T66a severe
isolate: II. Warm temperature, Valencia/macrophylla.
Unchallencred1/
Challencred
1/2/
CTV
No. of
Decline
index/
No. of
Decline
No. of
dead
isolate
plants
plants
index
plants
after
10 m<
challe
Tila
1
2.0
3
9
3/3
T26
4
1.7
5
4.2
0/5
T30
NE-/
NE
NE
NE
NE
T55a
2
2.0
2
2.2
0/2
Healthy 5
or control
plants uninoculated
with mild isolate
2.2
6
6.8
3/6
to
/ One year old plants were graft inoculated with leaf pieces under bark flaps on the
stem from donor plants infected with the indicated mild CTV isolates.
/ After verifying virus infection with mild isolates by DAS-ELISA with polyclonal
antibodies, the challenged plants were graft inoculated similarly with a T66a
severe isolate.
3 / . ...
Decline index is the average per treatment of the cumulative score given by visual
readings on three criteria: stem diameter, growth reduction, and foliage decline
symptoms, where 0 = healthy vigorous plant to 3 = severely diseased. Minimum index
= 0, maximum index =9. A high decline index sometimes was accompanied by plant
death.
4/
NE = treatment not evaluated.


96
Brlansky, R.H., Pelosi, R.R., Garnsey, S.M., Youtsey, C.O.,
Lee, R.F., Yokomi, R.K., and Sonoda, R.M. 1986. Tristeza
quick decline epidemic in South Florida. Proc. Fla. State
Hort. Soc. 99:66-69.
Calavan, E.C., Mather, S.M., and McEachern, E.H. 1978.
Registration, certification, and indexing of citrus
trees. Pages 185-222, in: Reuter, W., Calavan, E.C., and
Carman, E.G. (eds) The Citrus Industry. Vol IV. Crop
Protection. University of California. Riverside,
California.
Clark, M.F. and Adams, A.M. 1977. Characteristics of the
microplate method of enzyme-linked immunosorbent assay
for the detection of plant viruses. J. Gen. Virol.
34:475-483.
Clark, M.F., Lister, R.M., and Bar-Joseph, M. 1986. ELISA
techniques. Methods in Enzymology 118:742-767.
Cohen, M. 1972. A leaf insert graft used for virus
transmission in citrus. Pages 282-284, in: Price, W.C.
(ed) Proc. 5th Conf. Intern. Organ. Citrus Virol.
Gainesville, Florida.
Costa, A.S. and Muller, G.W. 1980. Tristeza control by cross
protection: A U.S.-Brazil cooperative success. Plant
Disease 64:538-531.
Davis, C.L. and Brlansky, R.H. 1991. Rapid detection of citrus
tristeza virus in citrus extracts using a gold labelled
antibody. J. Elect. Microsc. Tech. (Abstr., in press).
DeLange, J.H., Van Vuuren, S.P., and Bredell, G.S. 1981.
Groeipuntenting suiwer sitrusklone vir die
superplantskema van virusse. Subtropica 2 (5):11-16.
Fulton, R.W. 1986. Practices and precautions in the use of
cross protection for plant virus disease control. Ann.
Rev. Phytopathol. 24:67-81.
Garnsey, S.M. and Cambra, M. 1990. Enzyme linked immunosorbent
assays for citrus pathogens. In: Roistacher, C.N. (ed).
F.A.O. Citrus Disease Indexing Bulletin (in press).
Garnsey, S.M. and Jackson, J.L. 1975. A destructive outbreak
of tristeza in central Florida. Proc. Fla. State Hort.
Soc. 88:65-69.


LITERATURE CITED
Balaraman, K. and Ramakrishnan, K. 1980. Strain variation and
cross protection in citrus tristeza virus. Pages 60-68,
in: Calavan, E.C., Garnsey, S.M., and Timmer, L.W. (eds) .
Proc. 8th Conf. Intern. Organ. Citrus Virol. Riverside,
California.
Bar-Joseph, M. and Lee, R.F. 1990. Citrus tristeza virus.
Description of Plant Viruses No. 353 (No. 33 revised).
Commonwealth Mycological Institute/Association of Applied
Biologists. Kew Surrey, England. 7 p.
Bar-Joseph, M. and Loebenstein, G. 1973. Effects of strain,
source plant, and temperature on the transmissibility of
citrus tristeza virus by the melon aphid. Phytopathology
63:716-720.
Bar-Joseph, M. and Malkinson, M. 1980. Hen yolk as a source
of antiviral antibodies in the enzyme-linked
immunosorbent assay (ELISA): a comparison of two plant
viruses. J. Virol. Meth. 1:1-5.
Bar-Joseph, M., Garnsey, S.M., and Gonsalves, D. 1979a. The
closteroviruses: a distinct group of elongated plant
viruses. Adv. Virus Res. 25:93-168.
Bar-Joseph, M., Garnsey, S.M., Gonsalves, D., and Purcifull,
D.E. 1980. Detection of citrus tristeza virus. I. Enzyme-
linked immunosorbent assay (ELISA) and SDS-
immunodiffusion methods. Pages 1-8, in: Calavan, E.C.,
Garnsey, S.M., and Timmer, L.W. (eds). Proc. 8th Conf.
Intern. Organ. Citrus Virol. Riverside, California.
Bar-Joseph, M., Garnsey, S.M., Gonsalves, D., Moscovitz, M.,
Purcifull, D.E., Clark, M.F. and Loebenstein, G. 1979b.
The use of enzyme-linked immunosorbent assay for the
detection of citrus tristeza virus. Phytopathology
69:190-194.
Bar-Joseph, M., Gumpf, D.J., Dodds, J.A., Rosner, J.A., and
Ginzberg, I. 1985. A simple purification method for
citrus tristeza virus and estimation of its genome size.
Phytopathology 75:195-198.
94


35
(Cox et al. 1976; Fraser et al. 1968), against stem-pitting
isolates either in orange, grapefruit, and/or acid lime.
Several approaches have been reported for the evaluation of
mild isolates under greenhouse conditions. However, these
approaches have been addressed mostly to the evaluation of the
cross-protecting effect of mild isolates against stem pitting
and have included only the host reaction of Mexican lime,
sweet orange, or grapefruit seedlings (Roistacher et al. 1987,
1988; Van Vuuren and Noll, 1987). Another approach where the
challenge inoculations are made by using insect vectors to
screen mild isolates (Yokomi et al. 1987) has not been
extensively used.
The results of this work provide further evidence that
a) the cross-protection against the CTV-induced decline on
sweet/sour orange combinations may be possible; b) the
preliminary evaluation of mild isolates under greenhouse
conditions can be made in a relatively short time basis, and
c) the severe challenge isolate can be detected by using the
MCA-13 strain specific monoclonal antibodies. The recent
report of Miyakawa (1987) about the feasibility of cross
protection on sweet/sour orange combinations, supports these
conclusions.
Considering the relatively high virus titer found at cool
temperatures, it is advisable to propagate the donor plants
at temperatures in the range of 21-33C to better guarantee a
high percentage of CTV transmission to the receptor plants.


74
1053 3DF1 MCA-13
Fig. 4.4 Reaction of polyclonal antibodies no. 1053 and 3DF1
and MCA-13 monoclonal antibodies in dot-immunobinding assay
with twelve selected citrus tristeza virus (CTV) isolates.
Row A: 1 = Tila; 2 = T26; 3 = T30; 4 = T50a; 5 = T55a; 6 = T3;
7 = T4; 8 = T36. Row B: 1 = T62a; 2 = T65a; 3 = T66a; 4 =
T67a. Also in Row B are extract of healthy plants: 5 = Citrus
excelsa; 6 = Madam Vinous sweet orange; 7 = Duncan grapefruit;
8 = Mexican lime. Reaction conditions are described in
materials and methods.


Table 2.7 Effect of citrus tristeza virus (CTV) mild isolates on the development of
the CTV decline syndrome in plants unchallenged and challenged by the T66a severe
isolate: III. Cool temperature, Valencia/sour orange.
Unchallenged1/
Challenqed1/2/
CTV
No. of
Decline
index'
No. of
Decline
No. of dead
isolate
plants
plants
index
plants 10 months
after challenge
Tila
5
3.2
4
7.5
3/4
T26
1
1.0
2
5.5
0/2
T30
2
3.0
3
6.6
2/3
T55a
4
2.7
4
5.0
1/4
Healthy 1
or control
plants uninoculated
with mild isolate
3.0
2
7.5
1/2
/ One year
old plants were graft
inoculated
with leaf
pieces under bark flaps
stem from
i donor
plants infected
with the
indicated
mild CTV isolates.
2 / ,
After verifying virus infection with mild isolates by DAS-ELISA with polyclonal
antibodies, the challenged plants were graft inoculated similarly with a T66a
severe isolate.
to
o>
Decline index is the average per treatment of the cumulative score given by visual
readings on three criteria: stem diameter, growth reduction, and foliage decline
symptoms, where 0 = healthy vigorous plant to 3 = severely diseased. Minimum index
= 0, maximum index =9. A high decline index sometimes was accompanied by plant
death.
2/


ACKNOWLEDGEMENTS
I want to express my sincere gratitude and acknowledgment
to the following persons, institutions, and organizations, who
supported my graduate studies at the University of Florida.
The financial support from Consejo Nacional de Ciencia
y Tecnologa (CONACyT), as well as the support and leave of
absence from Instituto Nacional de Investigaciones Forestales
y Agropecuarias (INIFAP), both institutions from Mxico, is
greatly appreciated.
To my co-major professors, Drs. R.F. Lee and C.L.
Niblett, for their constant encouragement, guidance and
support throughout the course of the thesis work, and for
their personal interest in my academic preparation. To them
and the rest of the supervisory committee, Drs. S.M. Garnsey,
D.E. Purcifull, and R.K. Yokomi, for valuable suggestions in
revision of the manuscript.
The financial support from the Florida High Technology
and Industrial Council, and the Florida Citrus Production
Managers' Association, to carry out some parts of the thesis
work also is acknowledged.
I thank Drs. S.M. Garnsey and T.A. Permar, USDA Orlando,
and Drs. P. Moreno and M. Cambra, IVIC Valencia (Spain), for
supplying the MCA-13 and 3DF1 monoclonal antibodies,
respectively.
ii


CHAPTER 2
EVALUATION OF THE PROTECTING EFFECTS OF SOME MILD FLORIDA
ISOLATES OF CITRUS TRISTEZA VIRUS AGAINST THE DEVELOPMENT OF
THE DECLINE SYNDROME
Introduction
Cross protection is a control strategy used to reduce
losses due to plant viral diseases by the use of mild or
attenuated strains of a virus which prevent the effect or
expression of a related, and usually more severe, strain of
the same virus (Fulton, 1986). Cross protection can be useful
when the virus disease is endemic, causes great losses, and
no host genetic resistance is available (Fulton, 1986;
Gonsalves and Garnsey, 1989: Hamilton, 1985; Muller et al.
1982). Cross protection has been used commercially to control
CTV stem pitting isolates on Pera sweet orange in Brazil
(Costa and Muller, 1980; Muller, 1980), and is a part of South
Africa's citrus cultivar improvement program to reduce CTV-
induced stem pitting on grapefruit (DeLange et al. 1980;
Garnsey and Lee, 1988). Relatively little work has been done
to evaluate the potential of cross protection against the CTV-
induced decline (CTV-ID) on sour orange, as most countries
abandon sour orange as a rootstock when CTV-ID isolates become
prevalent. However, experiments conducted in Australia
6


Table 2.7
Table 2
Table 3
Table 3
Table 3
Table 3
Table 4
Table 4
Effect of citrus tristeza virus (CTV)
mild isolates on the development of the
CTV decline syndrome in plants
unchallenged and challenged by the
T66a severe isolate: III. Cool temperature,
Valencia/sour orange 26
8 Effect of citrus tristeza virus (CTV)
mild isolates on the development of the
CTV decline syndrome in plants
unchallenged and challenged by the
T66a severe isolate: IV. Cool temperature,
Valencia/macropylla 28
1 Transmission of citrus tristeza virus by
graft inoculation between selected
citrus hosts: I. Efficiency of leaf and
bark pieces as inoculum 45
2 Transmission of citrus tristeza virus by
graft inoculation between selected
citrus hosts: II. Overall rate of
transmission 46
3 Transmission of citrus tristeza virus by
graft inoculation between selected
citrus hosts: III. Effect of virus
isolates 48
4 Relative antigen titer of citrus
tristeza virus in different tissues of
Citrus excelsa and Madam Vinous sweet
orange host plants, as measured by enzyme-
linked immunosorbent assay 50
1 Relative sensitivity of DIBA, DAS-ELISA and
DAS-indirect with polyclonal (1053) and
monoclonal (3DF1 and MCA-13) antibodies
specific to citrus tristeza virus 71
2 Comparison of DIBA, DAS-ELISA and
DAS-indirect ELISA for the detection
of citrus tristeza virus (CTV) and
relative reactivity of polyclonal
(1053) and monoclonal (3DF1) (MCA-13)
antibodies with CTV isolates 75
x


Serological tests 42
Results 43
Graft transmission of citrus tristeza
virus isolates 43
Virus distribution and antigen
concentration in host tissues 47
Discussion 52
4. DEVELOPMENT OF A DOT-IMMUNOBINDING ASSAY FOR
CITRUS TRISTEZA VIRUS 57
Introduction 57
Materials and Methods 59
Antisera used 59
Sample preparation 60
Dot-immunobinding assay 61
DAS-ELISA 62
DAS-indirect ELISA 63
Evaluation 63
Evaluation of different buffers for
sample extraction 64
Results 65
Development of the dot-immunobinding assay 65
Evaluation 69
Evaluation of different buffers for
sample extraction 77
Discussion 77
5. SUMMARY AND CONCLUSIONS 86
LITERATURE CITED 94
BIOGRAPHICAL SKETCH 104
v


34
The relatively low decline index scores and low
occurrence of dead plants at both temperatures in the
unprotected control plants challenged with the T66a isolate,
indicates that under greenhouse conditions, the use of one
single challenge isolate might not be sufficient to obtain an
appropriate rate of decline in a short time basis. It is well
documented that CTV occurs naturally as mixtures of isolates
or strains with diverse biological properties (Garnsey et al.
1987, McClean, 1974). The T66a severe isolate was originally
isolated from an infected field source, and subsequently aphid
transmitted to avoid contamination with other viruses (Garnsey
et al. 1987; Yokomi and Garnsey, 1987) It is a possibility
that part of the original decline components from the field
source could have been lost in the subsequent aphid
transmissions. To overcome this possibility, it may be
advisable in the future to use a mixture of several severe
isolates as a challenge to enhance the possibility of
obtaining an appropriate occurrence of decline under
greenhouse conditions. Another alternative could be the use
of higher populations of aphids (50 or 100) to obtain a more
complete complex of CTV severe isolates from field samples.
Cross-protection using mild virus isolates as a strategy
to reduce losses due to CTV has been used in Brazil (Costa and
Muller, 1980; Muller, 1980), South Africa (DeLange et al.
1980; Garnsey and Lee, 1988), Japan (Ieki, 1989; Koizumi,
1986), India (Balaraman and Ramakrishnan, 1980) and Australia


Table 3.2 Transmission of citrus tristeza virus by graft inoculation between selected citrus
hosts: II. Overall rate of transmission.
Receptor host
Donor host
Madam Vinous
Mexican
lime
Grapefruit
Average
Citrus excelsa
72.4-/-/b^/
86.9
ab
60.7 b
72.5-/b
Mexican lime
93.1 a
76.9
b
89.3 a
85.2 ab
Madam Vinous
86.7 ab
100.0
a
84.6 a
90.6 a
4.
CTi
/ Percent transmission to plants with at least one inoculum piece (of four) alive 21 days
post-inoculation. Number indicates overall of transmission for all virus isolates
combinations.
2 / ,
Each value represents a minimum of 27 plants.
3 / ...
' Numbers in the same column following by different letters are statistically different
by Duncan's tests (P < 0.05).
4 / , ,
Number indicates the overall transmission for all receptor/virus isolate combinations.


TABLE OF CONTENTS
Page
ACKNOWLEDGMENTS
LIST OF FIGURES vi
LIST OF TABLES ix
ABSTRACT xi
CHAPTER
1. INTRODUCTION 1
2. EVALUATION OF THE PROTECTING EFFECTS OF SOME
FLORIDA ISOLATES OF CITRUS TRISTEZA VIRUS AGAINST
THE DEVELOPMENT OF THE DECLINE SYNDROME 6
Introduction 6
Materials and Methods 9
Virus isolates and donor hosts 9
Inoculation of CTV isolates and receptor hosts. 10
Serological tests 11
Detrimental effects of mild isolates and
evaluation of cross-protection 14
Results 14
Antigen titers of mild and severe CTV
isolates with polyclonal and MCA-13
monoclonal antibodies 14
Detrimental effect of mild isolates and
evaluation of cross-protection 21
Discussion 27
3. EFFECTIVENESS OF DIFFERENT CITRUS SPECIES AS
DONOR HOSTS FOR GRAFT TRANSMISSION OF CITRUS
TRISTEZA VIRUS 38
Introduction 38
Materials and Methods 40
Virus isolates and donor hosts 40
Grafting procedures and receptor hosts 41
Virus distribution and antigen
concentration in host tissues 41
Purification of CTV 42
iv


CHAPTER 5
SUMMARY AND CONCLUSIONS
Four different naturally occurring Florida mild citrus
tristeza virus (CTV) isolates were evaluated under greenhouse
conditions at two temperature regimes for their cross-
protecting ability against the induced decline by a CTV severe
challenge isolate in two susceptible scion/rootstock
combinations. DAS-ELISA with polyclonal antisera was used to
determine the total antigen titer in plants inoculated with
mild isolates and those challenged with the severe isolate.
The MCA-13 strain specific monoclonal antibody was
successfully used in DAS-indirect ELISA for differential
isolate detection and quantitation of the severe challenge
isolate in mixed infections.
The effectiveness of the graft transmission of CTV was
evaluated with three CTV isolates by using leaf and bark
tissue from three citrus donor hosts to three receptor hosts.
The distribution of the virus in different plant tissues was
also studied.
A dot-immunobinding assay (DIBA) was adapted for CTV
detection. The sensitivity level of DIBA was evaluated using
three different polyclonal and two monoclonal antibodies and
86


83
Garnsey, 1990; Permar et al. 1990), and it has been used for
strain discrimination of CTV-induced decline in several field
surveys (Irey, et al. 1988; Rocha-Pea et al. 1990; Yokomi et
al. 1990). The slight positive reaction found frequently in
DIBA with some CTV isolates with mild biological properties
occurred when the ascites fluid at a dilution of 1:5,000 was
incubated for 18 hr. These slight positive reactions could
be prevented by using a shorter (2 hr) incubation time or by
using the ascites fluid at a dilution of 1:20,000 and an
incubation time of 18 hr (Rocha-Pea et al. 1990) The
isolate T62a has been characterized previously as severe on
the basis of a severe vein flecking, leaf cupping and stunting
of Mexican lime plants, severe growth reduction on grapefruit
and Madam Vinous sweet orange seedlings, and moderate growth
reduction on sweet/orange combinations (Rocha-Pea and Lee,
unpublished). The weak reaction found with isolate T62a in
DIBA and its low OD405 when MCA-13 was used in DAS-indirect
ELISA (Table 4.2), suggests that some severe CTV isolates may
not be recognized by the MCA-13. That T62a was in high
concentration in these experiments is verified by DAS-ELISA
using polyclonal antibodies and DAS-indirect ELISA using the
3DF1 MCA (Table 4.2).
Three buffers, TBS, PBS and carbonate, with and without
0.05% Tween 20, were evaluated for their effects on the
sensitivity of DIBA, DAS-ELISA and DAS-indirect ELISA with
two antibodies and three CTV isolates. In DIBA, PBS gave the


12
severe CTV isolate T66a was detected in the challenged plants
by DAS-indirect ELISA using the MCA-13 strain specific
monoclonal antibody which reacts strongly against most severe
CTV isolates (Permar et al. 1990). For efficiency the
serological tests were performed on both experiments at the
same time.
Routinely, 0.5 g of bark, petioles and midribs of new,
fully expanded tissue were finely chopped with a razor blade
and ground, using a Tekmar Tissumizer, in 5 ml of phosphate
buffered saline (PBS)-Tween + polyvinyl pyrrolidone {PBS = 8
mM Na2HP04, 14 mM KH2P04, 15 mM NaCl, pH 7.4, (+ 0.1 % Tween
20 + 2% polyvinyl pyrrolidone (PVP-40 Sigma)}. Unless stated
otherwise, 200 microliters samples were used per well of the
microtiter plates and three washings with PBS-Tween (phosphate
buffered saline + 0.1 % Tween 20) were performed between
steps. The immunoglobulins (IgG) present in the whole CTV
antiserum were purified by the Protein A-Sepharose affinity
method (Miller & Stone, 1978). A portion of purified
immunoglobulins were conjugated to alkaline phosphatase by the
glutaraldehyde method (Clark et al. 1986). Polystyrene
Immulon II microtiter plates (Dynatech Laboratories) were
coated with 2.0 /g/ml of purified IgG in carbonate buffer
(0.015 M NaHC03, 0.03 M NaC03, pH 9.6) and incubated for 6 hr
at 37C. Antigen samples were added to the wells and
incubated for 18 hr at 5C. Enzyme conjugate was used at a
dilution of 1:1,000 in conjugate buffer (PBS-Tween + 2%


To my beloved wife Alicia, and to my adored son Eric,
whom are the world of my life.
To my parents Herminio and Dolores, and my brothers and
sisters.


65
yf-
with polyclonal IgG No. 1053 and the 3DF1 MCA as described
before.-^ Polyclonal IgG No. 1053 (1.0 /Ltg/ml) was incubated for
2 hr at 37C in 1% BSA, 2% PVP, TBS containing 1:200 (v/v)
buffer extract of bark from healthy Madam Vinous sweet orange,
prior to use in DIBA. Substrate reaction for DAS-ELISA and
DAS-indirect ELISA was quantified after 60 min.
Results
Development of the dot-immunobindinq assay. The
reactivity of each antibody and level of background on the
nitrocellulose membranes varied with each IgG used, IgG
concentration and incubation time. Reactions were affected
by the blocking solutions and buffers used for dilution of
the antibodies and the commercial goat anti-species IgG
conjugates.
Six different solutions were evaluated as blocking
agents. TBS alone (Fig. 4.1 A) and 10% horse serum (not
shown) did not prevent the nitrocellulose sheets from turning
dark; whereas 3% BSA, 3% gelatin, 0.5% non-fat dry milk, and
5% Triton X-100 all gave an acceptably white membrane with the
different polyclonal IgGs tested (Fig 4.1 B, C, D, E) The
3% gelatin blocking solution gave the best contrast between
the green color for healthy samples and different intensities
of a purple color for CTV infected samples (Fig. 4.1 C). The
use of Triton X-100 as a blocking agent partially removed the
green material from the nitrocellulose sheets; however, a


CHAPTER 4
DEVELOPMENT OF A DOT-IMMUNOBINDING ASSAY FOR
CITRUS TRISTEZA VIRUS
Introduction
Several serological methods have been developed and used
to diagnose the presence of citrus tristeza virus (CTV) in
infected tissue. These methods include: SDS-immunodiffusion
procedures (Garnsey et al. 1979; Bar-Joseph et al. 1980), in
situ immunofluorescence (Brlansky et al. 1984; Tsuchizaki et
al. 1978), serologically specific electron microscopy
(Brlansky et al. 1984; Garnsey et al. 1980), gold
immunolabeling microscopy (Davis and Brlansky, 1990), and
several enzyme-linked immunosorbent (ELISA) procedures (Bar-
Joseph and Malkinson, 1980; Bar-Joseph et al. 1979b) including
the use of the biotin-avidin system (Irey et al. 1988) and the
enzyme-amplified ELISA (Ben-Ze'ev et al. 1988) to increase the
sensitivity of detection. Each of these methods has different
advantages and sensitivity levels, and therefore, has been
used for different purposes and applications (Rocha-Pea and
Lee, 1990).
Polyclonal antisera specific for CTV have been developed
in different animal species, such as rabbits (Gonsalves et al.
1978; Tsuchizaki et al. 1978; R.F. Lee, unpublished;), and
57


93
sensitive as either ELISA procedure evaluated for CTV
diagnosis. The entire test could be performed in 2-3 hours
using polyclonal antibodies (slightly longer with monoclonal
antibodies), and minimal laboratory equipment was needed. The
use of small containers such as the polypropylene covers for
incubation and washing steps enabled the recovery and re-use
of both virus specific antibodies and goat anti-species IgG
conjugates for at least four weeks when the solutions were
properly preserved with 0.02% sodium azide and stored at 5C.
The template constructed from the rack of a micropipet tip
holder box enabled the uniform spacing of samples on the
nitrocellulose membranes.
One disadvantage of DIBA as compared to ELISA procedures
is the lack of quantitative measurements. However, with
appropriate positive and negative controls, DIBA can be used
reliably for routine diagnostic work where no quantitative
measurements are needed. When quantitation of CTV antigens
is required, ELISA should be used.


11
commercial potting mixture (Pro-mix BX) in five liter plastic
containers, and fertilized with a mixture of NPK (20-10-20)
every other week. Pest and disease management included the
application of 0.300 g active ingredient (a.i.)/plant of
aldicarb and 0.86 g a.i /L soil drench of ridomil twice a
year. The experiment was conducted in a greenhouse with mean
minimum and maximum temperatures of 21 and 38C, respectively.
At least 15 plants were inoculated for each CTV isolate/scion/
rootstock combination.
A second set of one-year old Valencia sweet orange plants
budded on sour orange and C. macrophvlla rootstocks were
inoculated with each of the CTV mild isolates and challenged
with the T66a severe isolate as previously described. These
plants were placed in a greenhouse with controlled mean
minimum and maximum temperatures of 21 and 33C, respectively,
to evaluate the effect of temperature on the cross protecting
ability of the CTV isolates. At least 10 plants were
inoculated per CTV isolate/scion/rootstock combination.
Fertilization and plant pest and disease management was as
above.
Serological tests. CTV infection and relative antigen
titer of inoculated plants were determined throughout the
study by the double antibody sandwich enzyme-linked
immunosorbent assay (DAS-ELISA) (Bar-Joseph et al. 1979b,
1980), using polyclonal antiserum No. 1053 prepared against
whole, unfixed CTV isolate T26 (R.F. Lee, unpublished). The


27
uninoculated healthy controls showed decline index scores of
5.0, 4.3, and 3.0, respectively. The plants pre-inoculated
with the T26 and T55a isolates and challenged with T66a showed
scores of 4.2 and 3.5, respectively. The uninoculated
controls challenged with T66a showed a decline index of 5.4
(Table 2.8). No dead plants were scored in this experiment
10 months after the challenge inoculations (Table 2.8).
Phloem necrosis at the bud union was not observed in any
treatment at either temperature at five or ten months after
the challenge inoculation (not shown).
Discussion
In this study four different naturally occurring Florida
mild CTV isolates were evaluated for their cross-protecting
ability against the development of the CTV-induced decline in
two susceptible scion/rootstock combinations. DAS-ELISA with
polyclonal antisera was used to determine the total antigen
titer in plants inoculated with mild isolates and those also
challenged with the severe T66a isolate. The MCA-13
monoclonal antibody was evaluated in DAS-indirect ELISA for
quantitation of the T66a severe challenge isolate in mixed
infections.
There were some limitations in the transmissibility of
CTV by leaf piece grafts from the different hosts used to
propagate the CTV isolates. At the beginning of the work most
of the CTV isolates had been propagated in Citrus excelsa
plants, which has been reported as an excellent propagation


Table 3.3 Transmission of citrus tristeza virus by graft inoculation between selected citrus
hosts: III. Effect of virus isolates.
Donor host
Virus
isolate
Citrus
excelsa
Mexican
lime
Madam Vinous
sweet orange
Averaae
T26
69.0-/-/a-/
89.3
a
92.8 a
83.5-/a
T30
76.7 a
65.2
b
100.0 a
80.7 a
T66a
71.4 a
96.7
a
77.3 b
83.5 a
. 03
/ Percent transmission to plants with at least one inoculum piece alive 21 days post
inoculation. Number indicates overall of transmission for all donor/receptor host
combinations.
2 / .
' Each value represents a minimum of 27 plants.
3 / ...
Numbers in the same column following by different letters are statistically different
by Duncan's tests (P < 0.05).
4 / ,
Number indicates overall transmission for all donor/receptor host combinations.


The technical assistance and friendship of N. Berger, S.
Marquardt, T. Nguyen, S. Jackson, and J. Zellers, as well as
the help of T. Zito in the photographic work, and M. Ahnger
with the statistical analysis, all at the Citrus Research and
Education Center, at Lake Alfred, is greatly appreciated.
The warm friendship I found in my fellow students, lab
technicians, administrative staff, and most faculty members
at the Plant Pathology Department, Gainesville, and at the
Citrus Research and Education Center, Lake Alfred, was indeed
encouraging and will be unforgettable.
Finally, I want to thank my wife Alicia for her constant
encouragement and patience to endure the hardship of my
pursuit of this graduate degree.
iii


100
McClean, A.P.D. 1957. Tristeza virus of citrus: Evidence for
absence of seed transmission. Plant Disease Reporter 41:
821.
McClean, A.P.D. 1974. The tristeza virus complex. Pages 59-
66, in: Weathers, L.G. and Cohen, M. (eds) Proc. 6th
Conf. Intern. Organ. Citrus Virol. Riverside, California.
Miyakawa, T. 1987. Protection against citrus tristeza seedling
yellows infection in citrus by pre-inoculation with stem
pitting isolates. Phytophylactica 19:193-196.
Miller, T.J. and Stone, H.O. 1978. The rapid isolation of
ribonuclease-free immunoglobulin G by Protein A-Sepharose
affinity chromatography. J. Immunol. Meth. 4:111-125.
Muller, G.W. 1980. Use of mild strains of citrus tristeza
virus (CTV) to re-establish commercial production of
''Pera sweet orange in Sao Paulo, Brazil. Proc. Fla.
State Hort. Soc. 93:62-64.
Muller, G. W. and Costa, A.S. 1977. Tristeza control in Brazil
by preimmunization with mild strains. Proc. Intern. Soc.
Citriculture 3:868-872.
Muller, G.W. and Costa, A.S. 1987. Search for outstanding
plants in tristeza infected orchards: The best approach
to control the disease by preimmunization. Phytohylactica
19:197-198.
Muller, G.W. and Garnsey, S.M,. 1984. Susceptibility of citrus
varieties, species, citrus relatives, and non-rutaceous
plants to slash-cut mechanical inoculation with citrus
tristeza virus (CTV). Pages 33-40, in: Garnsey, S.M.,
Timmer, L.W., and Dodds, A.J. (eds) Proc. 9th Conf.
Intern. Organ. Citrus Virol. Riverside, California.
Muller, G.W., Rezende, J.A.M., and Costa, A.S. 1982.
Preimmunization: An approach to control plant virus
diseases. Proc. 1st Conf. Impact Viral Dis. Rio de
Janeiro, Brazil.
Norman, G., Price, W.C., Grant, T.J., and Burnett, H. 1961.
Ten years of tristeza in Florida. Proc. Fla.State Hort.
Soc. 74:107-111.
Permar, T.A. and Garnsey, S.M. 1990. Comparison of biological
indexing and immunological assays for identifying severe
Florida isolates of citrus tristeza virus. In: Proc.
11th Conf. Intern. Organ. Citrus Virol. Riverside,
California (in press).


97
Garnsey, S.M. and Lee, R.F. 1988. Tristeza. Pages 48-50, in:
Whiteside, J.O., Garnsey, S.M., and Timmer, L.W. (eds).
Compendium of Citrus Diseases. APS Press. 80 p.
Garnsey, S.M. and Muller, G.W. 1988. Efficiency of mechanical
transmission of citrus tristeza virus. Pages 46-54, in:
Timmer, L.W., Garnsey, S.M., and Navarro. L. (ed). Proc.
10th Conf. Intern. Organ. Citrus Virol. Riverside,
California.
Garnsey, S.M. and Whidden, R. 1970. A rapid technique for
making leaf tissue grafts to transmit citrus viruses.
Plant Dis. Rep. 54: 907-908.
Garnsey, S.M., Bar-Joseph, M., and Lee, R.F. 1981a.
Applications of serological indexing to develop control
strategies for citrus tristeza virus. Proc. Intern. Soc.
Citriculture 1:448-452.
Garnsey, S.M., Christie, R.G., Derrick, K.S., and Bar-Joseph,
M. 1980a. Detection of citrus tristeza virus. II. Light
and electron microscopy of inclusions and viral
particles. Pages 9-16, in: Calavan, E.C., Garnsey, S.M.,
and Timmer, L.W. (eds). Proc. 8th Conf. Intern. Organ.
Citrus Virol. Riverside, California.
Garnsey, S.M., Gonsalves, D., and Purcifull, D.E. 1977.
Mechanical transmission of citrus tristeza virus.
Phytopathology 67:965-968.
Garnsey, S.M., Gonsalves, D., and Purcifull, D.E. 1979. Rapid
diagnosis of citrus tristeza virus infections by sodium
dodecyl sulphate-immunodiffusion procedures.
Phytopathology 69:88-95.
Garnsey. S.M., Gumpf, D.J., Roistacher, C.N. Civerolo, E.L.,
Lee, R.F., and Yokomi, R.K. 1987. Toward a standardized
evaluation of the biological properties of citrus
tristeza virus. Phytophylactica 19:151-157.
Garnsey, S.M., Lee, R.F., and Brlansky, R.H. 1981b.
Preparation and stability of infectious citrus tristeza
virus (CTV). Phytopathology 71:218 (Abstr.).
Garnsey, S.M., Lee, R.F., Youtsey, C.O., Brlansky, R.H., and
Burnett, H.C.. 1980b. A survey for citrus tristeza virus
in registered budwood sources commercially propagated on
sour orange rootstock in Florida. Proc. Fla. State Hort.
Soc. 93:7-9.


3
(Brlansky et al. 1984; Tsuchizaki et al. 1978). Each of these
methods has different advantages, disadvantages, and
sensitivity levels. Therefore, a certain method may be used
for a particular purpose. Some methods, such as ELISA, are
dependable and widely used for indexing purposes (Garnsey et
al. 1981a).
Recently a monoclonal antibody was developed against a
decline-inducing isolate from Florida which, in ELISA, reacted
specifically with several severe CTV isolates from diverse
geographical areas, but not with mild isolates from the same
areas (Permar et al. 1990).
Control of CTV is difficult. In those few areas of the
world where CTV still is not present, quarantine and virus-
free certification programs are maintained to prevent the
introduction of infected budwood sources (Bar-Joseph et al.
1983, 1989). Likewise, in those areas with low CTV incidence,
large scale surveys and suppression measures are carried out
to reduce disease spread to other trees and locations and
prolong the use of sour orange as a rootstock (Bar-Joseph et
al. 1989). Once the disease becomes endemic, two situations
can result: a) CTV-induced decline develops and kills plants
grafted onto sour orange rootstock, whereas tolerant
rootstocks do not decline, and/or b) CTV-stem pitting can
affect sweet orange and/or grapefruit scions regardless of the
rootstock, resulting in a loss of plant vigor and yield. Mild
strain cross protection is the only known control measure


89
Plants pre-inoculated with the Tila mild isolate, and
further challenged with the T66a severe isolate showed the
highest occurrence of decline, reflected in high decline index
scores in the whole experiment, indicating apparently that the
combination of both Tila and T66a isolates produced a more
severe reaction in the challenged plants even than that caused
by the T66a isolate alone in the unprotected control plants.
There was a relatively low occurrence of decline at both
temperatures in the unprotected control which were challenged,
indicating that to better guarantee an appropriate occurrence
of decline symptoms under greenhouse conditions, it would be
advisable toe use a mixture of more than one severe isolate
as the challenge.
The results of this work provide further evidence that:
a) the cross-protection against the induced decline on
sweet/sour orange combinations is possible, and it can be
evaluated preliminarily under greenhouse conditions in a
relatively short time period of 18-24 months. This time period
estimate includes from six to twelve months to infect the test
plants with mild isolates and verification of infection by
serology, two months for challenge inoculation with the severe
isolate and ten months for final evaluation; b) The severe
challenge isolate can be differentially detected from the mild
isolates by using the MCA-13 strain specific monoclonal
antibody; c) Mild isolates can be evaluated without the risk


19
Table 2.3 Relative antigen titer of citrus tristeza virus
(CTV) mild isolates in plants unchallenged and challenged by
the T66a severe isolate: III. Cool temperature, Valencia/sour
orange.
Unchallenaed1/
Challenaed1/2/
CTV
ODac3/
OD
isolate
Polyclonal
MCA-13
Polyclonal
MCA-13
Tila
0.404^
2
0.060
a
0.548
a
0.477 a
T26
0.405
a
0.078
a
0.195
b
0.055 c
T30
0.138
a
0.054
a
0.174
b
0.047 c
T55a
0.308
a
0.055
a
0.276
b
0.057 c
Healthy
0.039
b
0.016
a
0.292
b
0.283 b
or control
plants uninoculated
with mild isolate
One-year-old plants were graft inoculated with leaf
pieces under bark flaps on the stem from donor plants
infected with the indicated mild CTV isolates.
After verifying virus infection with mild isolates by
DAS-ELISA with polyclonal antibodies, the challenged
plants were graft inoculated similarly with the T66a
severe isolate.
3/ Optical density at 405 nm (OD405) was measured after 120 min
of reaction.
Serological detection was carried out by DAS-ELISA with
polyclonal antisera and DAS-indirect ELISA with MCA-13
severe strain specific monoclonal antibody six months
after inoculation. Value is the average of duplicate
assays of the corresponding plants in Table 2.7.
Numbers in the same column followed by different letters
are statistically different by Duncan's test (P < 0.05).


66
Fig. 4.1 Effect of different blocking solutions on the
reaction of polyclonal antibodies no. 1053 in dot-
immunobinding assay with citrus tristeza virus (CTV) isolates:
A. TBS alone; B. 3% bovine serum albumin (BSA); C. 3% gelatin;
D. 0.5% non-fat dry milk; E. 5% Triton X-100. Number 1, CTV
T66a; 2, CTV T26; 3, buffer extract from healthy sweet orange
plants; 4, TBS-Tween. Reaction conditions are described in
materials and methods.


45
Table 3.1 Transmission of citrus tristeza virus by graft
inoculation between selected citrus hosts: I. Efficiency of leaf
and bark pieces as inoculum.
Inoculum
tissue
Inoculum
survival%-/
% plants with
at least one % transmission-7
successful graft
leaf
8 3. O^a^7
92.0
89.2 a
bark
66.0 b
90.0
75.6 b
Measured at 21 days post-inoculation.
Percent transmission to plants with at least one inoculum
piece (of four) alive, measured serologically by DAS-ELISA at
3 and 5 months post-inoculation. Number indicates overall
transmission for all donor/receptor/virus isolate combinations.
A total of 270 plants (135 each) were inoculated with four pieces
of either leaf or bark tissue. Number indicates overall survival
for all donor/receptor/virus isolate combinations.
Numbers in the same column followed by different letters are
statistically different by Duncan's test (P < 0.05).


20
Table 2.4 Relative antigen titer of citrus tristeza virus
(CTV) mild isolates in plants unchallenged and challenged by
the T66a severe isolate: IV. Cool temperature,
Valencia/macrophylla.
Unchallenged17 Challenged1727
CTV OD,0S^
isolate
Polyclonal
MCA-13
Polyclonal
MCA-
13
Tila
NE^7
NE
NE
NE
T26
0.36 l-7a-7
0.049 a
0.406
a
0.166
a
T30
NE
NE
NE
NE
T55a
0.335 a
0.075 a
0.456
a
0.320
a
Healthy
0.036 b
0.013 b
0.432
a
0.421
a
or control
plants uninoculated
with mild isolate
-7 One-year-old plants were graft inoculated with leaf
pieces under bark flaps on the stem from donor plants
infected with the indicated mild CTV isolates.
27 After verifying virus infection with mild isolates by
DAS-ELISA with polyclonal antibodies, the challenged
plants were graft inoculated similarly with the T66a
severe isolate.
-7 Optical density at 405 nm (OD405) was measured after 120 min
of reaction.
-7 NE = treatment not evaluated.
-7 Serological detection was carried out by DAS-ELISA with
polyclonal antisera and DAS-indirect ELISA with MCA-13
severe strain specific monoclonal antibody six months
after inoculation. Value is the average of duplicate
assays of the corresponding plants in Table 2.8.
Numbers in the same column followed by different letters
are statistically different by Duncan's test (P < 0.05).


I certify that I have read this study and that in my opinion,
it conforms to acceptable standards of scholarly presentation and
is fully adequate, in scope and quality, as a dissertation for the
deqree of Doctor of Philosophy.
December 1990
Deant
Agriculture
ollege of^
Dean, Graduate School


10
clearing, stunting and stem pitting in Mexican lime seedlings
and severe decline on sweet/sour orange combinations (Garnsey
et al. 1987; Yokomi et al. 1987). The CTV isolates were
propagated in either C. excelsa Wester, Madam Vinous sweet
orange or Mexican lime plants and maintained in a greenhouse
with mean minimum and maximum temperatures of 21 and 38C,
respectively. Inoculum source tissue from donor hosts was
evaluated by serological indexing (see below) to confirm the
presence of CTV before being used as inoculum.
Inoculation of CTV isolates and receptor hosts. One-
year-old Valencia sweet orange plants budded on either sour
orange or C. macrophvlla Wester rootstocks, were graft
inoculated in the stem with each of the CTV mild isolates
using three leaf pieces or blind buds per plant (Garnsey and
Whidden, 1970; Garnsey et al. 1987). Inoculum tissue was
sealed firmly into the receptor stems with plastic grafting
tape. Three weeks later the grafting tape was removed and
the plants were evaluated for survival of grafted tissue and
reinoculated if the grafted tissue had not survived. After
verifying by serological indexing that infection by mild
isolates had taken place, a minimum of four inoculum pieces
of T66a infected tissue were used to challenge the test
plants, and they were reinoculated if at least two inoculum
pieces were not alive 21 days post-challenge. Surviving
inoculum tissue was left in place for the duration of the
experiment. Inoculated receptor plants were grown in a


absorbed with buffer extract of healthy
plant. B. Polyclonal 1053 (1.0 /g/ml)
incubated with 1:200 (v/v) buffer extract
from healthy plants for 2 hr at 37C
prior to use. C and D, monoclonal 3DF1
(1.0 nq/ml) and MCA-13 (1:5,000 dilution)
antibodies not cross-absorbed with healthy
extract. Number 1, CTV T66a; 2, CTV T26;
3, buffer extract from healthy sweet orange
plants; 4, TBS-Tween. Reaction conditions
are described in materials and methods 68
Figure 4.3 Relative sensitivity level of different
polyclonal and monoclonal antibodies
specific to citrus tristeza virus (CTV)
in dot-immunobinding assay. Rows A, B and C,
polyclonal antibodies nos. 1051, 1052, and
1053, respectively. Rows D and E,
monoclonal antibodies 3DF1 and MCA-13,
respectively. Extract of Citrus excelsa
greenhouse grown plants infected with
CTV T36 isolate was prepared in
TBS-Tween 1:10 (w/v) and two fold
diluted with extract of healthy plants.
Reaction conditions are described in
materials and methods 70
Figure 4.4 Reaction of polyclonal antibodies no.
1053 and 3DF1 and MCA-13 monoclonal
antibodies in dot-immunobinding assay
with twelve selected citrus tristeza
virus (CTV) isolates. Row A: 1 = Tila;
2 = T26; 3 = T30; 4 = T50a; 5 = T55a;
6 = T3; 7 = T4; 8 = T36. Row B. 1 = T62a;
2 = T65a; 3 = T66a; 4 = T67a. Also in Row B
are extracts of healthy plants: 5 = Citrus
excelsa; 6 = Madam Vinous sweet orange;
7 = Duncan grapefruit; 8 = Mexican lime.... 74
Figure 4.5 Evaluation of different extraction
buffers on the sensitivity of DIBA with
citrus tristeza virus (CTV) isolates T26,
T62a and T66a. TBS = Tris buffered saline,
TBST = TBS containing 0.05% Tween, PBS =
phosphate buffered saline, PBST = PBS +
0.05% Tween, Carb = carbonate buffer, CarbT
= Carb + 0.05% Tween, HC = healthy control.
The IgG of polyclonal antibodies No. 1053
or monoclonal antibody 3DF1 were used as
primary antibodies followed by the goat anti
rabbit and anti- mouse IgG conjugates,
respectively. Samples were 2 ;xl of buffer
vii


13
polyvinyl pyrrolidone + 0.2% bovine serum albumin) and
incubated for 6 hr at 37C. The reaction with one mg/ml of p-
nitrophenyl phosphate (Sigma) in 10% triethanolamine, pH 9.8,
was measured after 120 min at 405 nm (OD405) with a
Labinstruments model EAR 400 AT ELISA plate
spectrophotometer. Samples were considered positive when OD405
values were higher than 0.100 or three times the mean of
healthy controls, whichever was greater.
For DAS-Indirect ELISA, the microtiter plates were first
coated with IgG from antiserum No. 1053. Antigen samples were
added as described for DAS-ELISA. The MCA-13 strain specific
monoclonal antibody (hereafter MCA-13), as ascites fluid, was
added at a dilution of 1:5,000 (v/v) in conjugate buffer and
incubated 4 hr at 37C. After washing, goat anti-mouse IgG
labeled with alkaline phosphatase (Promega) at a dilution of
1:7,500 (v/v) in conjugate buffer was added and incubated for
2 hr at 37C. The enzyme reaction was carried out as for DAS-
ELISA.
For all serological tests, two replications were used
per sample. Positive controls included four mild isolates
(Tila, T26, T30, and T55a) and one severe (T66a) CTV isolate.
Negative controls included extraction buffer, and similar
buffer extracts from healthy C. excelsa and Valencia sweet
orange plants. A standard curve prepared with purified CTV
T26 isolate diluted to OD260 values of 0.04, 0.02, 0.01, 0.005,
0.0025, 0.0012, 0.0006, and 0.0003 diluted in buffer extract


Table 2.
the CTV
isolate:
8 Effect of citrus tristeza virus (CTV) mild
decline syndrome in plants unchallenged and
IV. Cool temperature, Valencia/macrophylla.
isolates on the development of
challenged by the T66a severe
Unchallenged1/
Challenged1/2/
CTV
No. of
Decline
index/
No. of
Decline
No. of dead
isolate
plants
plants
index
plants 10 months
after challenge
Tila
NE-/
NE
NE
NE
NE
T26
4
5.0
5
4.2
0/5
T30
NE
NE
NE
NE
NE
T55a
3
4.3
4
3.5
0/4
Healthy 3
or control
3.0
5
5.4
0/5
plants uninoculated
with mild isolate
CO
/ One year old plants were graft inoculated with leaf pieces under bark flaps on the
stem from donor plants infected with the indicated mild CTV isolates.
-/ After verifying virus infection with mild isolates by DAS-ELISA with polyclonal
antibodies, the challenged plants were graft inoculated similarly with a T66a
severe isolate.
3 /
' Decline index is the average per treatment of the cumulative score given by visual
readings on three criteria: stem diameter, growth reduction, and foliage decline
symptom, where 0 = healthy vigorous plant to 3 = severely diseased. Minimum index
= 0, maximum index =9. A high decline index sometimes was accompanied by plant
death.
4/
NE = treatment not evaluated.


LIST OF FIGURES
Page
Figure 2.1 Plot of purified citrus tristeza
virus (CTV) against optical density.
Bark of healthy Citrus excelsa
(0.5 g) tissue was ground in 5.0 ml
of phosphate buffered saline, pH 7.6,
+ 0.05% Tween + 2% polyvinyl pyrrolidone
and mixed with purified CTV T26 isolate
to give the desired optical density
at 260 nm (OD260) DAS-ELISA was
performed as described in materials
and methods. An extinction
coefficient of 2.0 was assumed
(Gonsalves et al. 1978) to estimate
the relative virus concentration 22
Figure 3.1 Plot of purified citrus tristeza
virus (CTV) against optical density.
Bark of healthy Citrus excelsa
(0.25 g) tissue was ground in 5.0 ml
of phosphate buffered saline, pH 7.6,
+ 0.05% Tween + 2% polyvinyl pyrrolidone
and mixed with purified CTV T26 isolate
to give the desired optical density
at 260 nm (OD260) DAS-ELISA was
performed as described in materials
and methods. An extinction
coefficient of 2.0 was assumed
(Gonsalves et al. 1978) to estimate
the relative virus concentration 51
Figure 4.1 Effect of different blocking solutions
on the reaction of polyclonal antibodies
no. 1053 in dot-immunobinding assay with
citrus tristeza virus (CTV) isolates:
A. TBS alone; B. 3% bovine serum albumin
(BSA); C. 3% gelatin; D. 0.5% non-fat dry
milk; E. 5% Triton X-100. Number 1,
CTV T66a; 2, CTV T26; 3, buffer extract
from healthy sweet orange plants;
4, TBS-Tween. Reaction conditions are
described in materials and methods 66
Figure 4.2 Reaction of polyclonal and monoclonal
antibodies in dot-immunobinding assay
with citrus tristeza virus (CTV) isolates.
A. Polyclonal 1053 (1.0 ig/ml) not cross-
vi


30
very low OD405 value could cause the whole treatment average to
be lower than 0.100. It has been suggested that a high titer
in a plant infected with a mild CTV isolate may be a relative
estimate of the protecting ability of mild isolates in cross
protection experiments (Koizumi and Kuhara 1984; Lee et al.
1987a).
Some differences were found in the reaction of the MCA-
13 monoclonal antibody in the different treatments and
temperatures evaluated. At warm temperatures plants pre
inoculated with mild isolates but unchallenged, gave low OD405
values in the range of the uninoculated control plants (Table
2.1 and 2.2). Likewise, at cool temperatures, the OD405 values
obtained with the MCA-13 with mild isolates were slightly
higher than those obtained at warm temperatures (Tables 2.3
and 2.4). This could be interpreted that the MCA-13
monoclonal antibody may react to some extent with mild
isolates when they are above a certain titer in the plants.
However, the OD405 values were always lower than 0.100, which
was considered a negative reaction.
When plants pre-inoculated with mild isolates and further
challenged with the T66a severe isolate were analyzed with the
MCA-13 in DAS-indirect ELISA (Tables 2.1 and 2.2), the OD405
values were generally lower than the unprotected challenged
control plants, even though the differences usually were not
statistically significant (with the exception T55a in Table
2.2) At cool temperatures, a similar phenomenon was observed


37
possibility that a lack of transmissibility by leaf piece
grafts of the CTV challenge isolate, can be interpreted
erroneously as protecting effect by mild isolates. On the
other hand, the inoculum tissue with the severe isolate is
left in place to enhance the probability of graft-transmission
in the challenged plants. This would supply a permanent
source of the severe isolate against the mild isolates which
may provide a stronger challenge pressure than happens under
natural conditions. If this occurs, mild isolates with
potential protecting ability under natural challenge
conditions could be underestimated or overlooked.


16
Table 2.1 Relative antigen titer of citrus tristeza virus
(CTV) mild isolates in plants unchallenged and challenged by
the T66a severe isolate: I. Warm temperature, Valencia/sour
orange.
Unchallenged17 Challenged1727
CTV OD,0_
isolate
Polyclonal
MCA-
13
Polyclonal
MCA-:
13
Tila
0.101-7afc£7
0.024
a
0.217
a
0.175
a
T26
0.123
ab
0.025
a
0.130
a
0.145
a
T3 0
0.091
ab
0.018
ab
0.212
a
0.189
a
T55a
0.145
a
0.014
b
0.194
a
0.181
a
Healthy
0.039
b
0.011
b
0.174
a
0.214
a
or control
plants uninoculated
with mild isolate
One-year-old plants were graft inoculated with leaf
pieces under bark flaps on the stem from donor plants
infected with the indicated mild CTV isolates.
27 After verifying virus infection with mild isolates by
DAS-ELISA with polyclonal antibodies, the challenged
plants were graft inoculated similarly with the T66a
severe isolate.
-7 Optical density at 405 nm (OD405) was measured after 120 min
of reaction.
-7 Serological detection was carried out by DAS-ELISA with
polyclonal antisera and DAS-indirect ELISA with MCA-13
severe strain specific monoclonal antibody six months
after inoculation. Value is the average of duplicate
assays of the corresponding plants in Table 2.5.
-7 Numbers in the same column followed by different letters
are statistically different by Duncan's test (P < 0.05).


42
study the relative distribution and antigen concentration of
the virus in different tissues of the host plant. Bark,
petioles, midribs, and leaf blades of four individual branches
of each test plant, were assayed individually by DAS-ELISA.
At least four replications were assayed for every host/virus
isolate combination tested.
Purification of CTV. Citrus tristeza virus was purified
from tender new tissue of C. excelsa greenhouse grown plants
infected with the T26 isolate, by the Driselase method
(Garnsey et al. 1981b; Lee et al. 1988b). The final virus
preparations were adjusted with 0.05 M Tris buffer to optical
density values (OD260) of 0.4 and stored in one ml aliquots at
18C.
Serological tests. The double antibody sandwich enzyme-
linked immunosorbent assay (DAS-ELISA) (Bar-Joseph et al.
1979b, 1980) was conducted with polyclonal antiserum no. 1053
prepared against whole, unfixed CTV isolate T26 (R.F. Lee,
unpublished). Polystyrene Immulon II microtiter plates
(Dynatech Laboratories) were used. Unless otherwise stated,
200 microliters were used per well of the microtiter plates
and, three washings with phosphate buffered saline (PBS)-Tween
{PBS = 8 mM Na2HP04, 14 mM KH2P04, 15 mM NaCl, pH 7.4, (+0.1
% Tween 20)> were performed between steps. Host tissue (bark,
petioles, midribs, etc.) was chopped finely with a razor blade
and ground in a Tekmar Tissumizer in extraction buffer (PBS-
Tween + 2% polyvinyl pyrrolidone (PVP-40 Sigma) at a 1:20


LIST OF TABLES
Table 2.
Table 2.
Table 2.
Table 2.
Table 2.
Table 2.
Page
1 Relative antigen titer of citrus
tristeza virus (CTV) mild isolates
in plants unchallenged and
challenged by the T66a severe
isolate: I. Warm temperature,
Valencia/sour orange 16
2 Relative antigen titer of citrus
tristeza virus (CTV) mild isolates
in plants unchallenged and
challenged by the T66a severe
isolate: II. Warm temperature,
Valencia/macrophylla 17
3 Relative antigen titer of citrus
tristeza virus (CTV) mild isolates
in plants unchallenged and
challenged by the T66a severe
isolate: III. Cool temperature,
Valencia/sour orange 19
4 Relative antigen titer of citrus
tristeza virus (CTV) mild isolates
in plants unchallenged and
challenged by the T66a severe
isolate: IV. Cool temperature,
Valencia/macrophylla 20
5 Effect of citrus tristeza virus (CTV)
mild isolates on the development of the
CTV decline syndrome in plants
unchallenged and challenged by the
T66a severe isolate: I. Warm temperature,
Valencia/sour orange 23
6 Effect of citrus tristeza virus (CTV)
mild isolates on the development of the
CTV decline syndrome in plants
unchallenged and challenged by the
T66a severe isolate: II. Warm temperature,
Valencia/macrophylla 25
ix


92
reactivity of each antibody with the different CTV isolates
tested. All polyclonal antibodies (nos. 1051, 1052, and 1053)
-
reacted similarly in DIBA, but all had to be cross-absorbed
with buffer extracts of healthy plants before use. All
polyclonals gave a strong positive reaction with the 12 CTV
isolates tested. The dilution end points of extracts from
CTV-infected samples were 1/320 when tested against the
polyclonal antibodies.
The 3DF1 monoclonal antibody, at a concentration of 1.0
ig/ml,reacted moderately with most of the CTV isolates tested.
Dilution end points of extracts from CTV-infected tissue were
1/320 in DIBA and 1/680 in DAS-indirect ELISA. The strongest
reactions in DIBA were obtained when the incubation time was
extended to 18 hr.
The MCA-13 monoclonal antibody used, at a dilution of
1:5,000 in DIBA, reacted strongly with CTV isolates T3, T36,
T65a, T66a, and T67a all sharing common severe biological
properties. However, there was a slight positive reaction
with two CTV isolates, T50a and T55a, having mild biological
properties and with one isolate (T62a) known to have severe
properties. The highest dilution end point of CTV-infected
extracts which reacted with MCA-13 in DIBA and DAS-indirect
ELISA was 1/320 using an incubation time of 18 hr.
There are several advantages to using DIBA over
conventional DAS-ELISA or DAS-indirect ELISA for CTV
detection. DIBA was rapid and easy to perform and, it was as


78
si I I
4>
*-
a
TBS
4


TBST



PBS

o

PBST


%
Carb



CarbT
O


H20



o
O
1053
3DF1
Figure 4.5 Evaluation of different extraction buffers on the
sensitivity of DIBA with citrus tristeza virus (CTV) isolates
T26, T62a and T66a. TBS = Tris buffered saline, TBST = TBS
containing 0.05% Tween 20, PBS = phosphate buffered saline,
PBST = PBS + 0.05% Tween, Carb = carbonate buffer, CarbT =
Carb + 0.05% Tween, HC = healthy control. The IgG of
polyclonal antibodies No. 1053 or monoclonal antibody 3DF1
were used as primary antibodies followed by the goat anti
rabbit and anti-mouse IgG conjugates, respectively. Samples
were 2 /I of buffer extracts of greenhouse-grown Madam Vinous
sweet orange plants infected with the CTV isolates ground at
a 1:10 dilution.


22
E
c
ID
O
*->
C\3
>.
*->
CO
c
Q)
Q
o
4
Q.
o
Purified virus (Optical Density 260 nm)
Figure 2.1 Plot of purified citrus tristeza virus (CTV)
against optical density. Bark of healthy Citrus excelsa (0.5
g) tissue was ground in 5.0 ml of phosphate buffered saline,
pH 7.6, + 0.05% Tween + 2% polyvinyl pyrrolidone and mixed
with purified CTV T26 isolate to give the desired optical
density at 260 nm (OD260) DAS-ELISA was performed as described
in materials and methods. An extinction coefficient of 2.0
was assumed (Gonsalves et al. 1978) to estimate the relative
virus concentration.
OJb /s


7
(Thornton et al. 1980), United States (Wallace and Drake,
1976; Yokomi et al. 1990), and Japan (Miyakawa, 1987) indicate
that cross protection against CTV-ID isolates may be possible.
The classical approach for selecting potential cross
protecting mild CTV isolates has been empirical, that is, by
collecting a number of CTV isolates from outstanding trees in
groves severely affected by the disease (Balaraman and
Ramakrishnan, 1980; Muller and Costa, 1987). Those isolates
are propagated on several scion-rootstock combinations in
nursery plants and further evaluated over a period of several
years under field conditions (Muller, 1980; Muller and Costa,
1977) This requires the handling and care of large numbers
of plants, extensive field space, and considerable time; the
results are evaluated over a period of 5-12 years. By this
approach, only six mild CTV isolates out of 45 mild isolates
originally selected in Brazil were useful for cross protection
(Costa and Muller, 1980).
Another approach for the selection of mild CTV isolates
for cross protection relies upon host effects for the strain
segregation of field collected severe CTV isolates (Roistacher
et al. 1988). The selection of potential cross protecting CTV
isolates was made from either symptomless or recovered
infected plants after extensive passage of the virus through
a series of different citrus and non-citrus hosts in the
greenhouse (Roistacher et al. 1987, 1988). This approach
provided nine mild or attenuated CTV isolates with outstanding


80
the commercial goat anti-species IgG conjugates, the
evaluation of some CTV specific polyclonal and monoclonal
antibodies and the testing of the effect of different
extraction buffers with and without Tween 20.
In some other systems where nitrocellulose membranes have
been used for the immunological detection of proteins either
by DIBA (Hibi and Saito, 1985; Powell, 1987; Graddon and
Randles, 1986) or by western blotting (Spinola and Cannon,
1985), some differences have been reported in the suitability
of different agents and buffers used for membrane blocking or
as diluents for both the virus specific antibodies and the
commercial goat anti-species IgG conjugates. In this study,
TBS, 10% horse serum, 3% BSA, 3% gelatin, 0.5% non-fat dry
milk and 5% Triton X-100 were evaluated as blocking agents.
The latter four were found to give an adequately white
background on the nitrocellulose membranes to permit
discrimination between infected and healthy samples. However,
3% gelatin was used routinely as it gave the most suitable
contrast between a green color for the healthy samples and a
distinct purple color for the infected samples.
It has been well documented that the presence of albumin
(BSA or egg ovalbumin) enhances the reactivity of antibodies
in diverse serological tests (Clark et al. 1986; Purcifull and
Batchelor, 1977). Likewise, PVP has been commonly used in
both extraction and conjugate buffers to prevent non-specific
reactions (Clark et al. 1986). In this study, a concentration


31
(Tables 2.3 and 2.4). In this situation, the T26 and T30
isolates generally gave the lowest OD405 values of all
treatments evaluated. This suggests that the mild isolates,
especially T26 and T30, but also T55a to some degree,
prevented or reduced the multiplication of the T66a challenge
isolate. Thus these mild isolates were apparently working as
cross-protecting agents.
These results provide further evidence of the usefulness
of the MCA-13 monoclonal antibody to detect the presence of
severe CTV isolates in mixed infections. The MCA-13 has been
previously used in other studies to evaluate the presence of
severe isolates in field cross-protection experiments (Rocha-
Pea et al. 1990; Yokomi et al. 1990). Determination of the
OD405 readings when the MCA-13 monoclonal antibody is used to
detect the presence of the severe isolate in cross-protection
experiments provides a measurable parameter to estimate the
ability of mild isolates to prevent the establishment of
severe challenge isolates in such experiments.
Some differences were found in the effects of CTV mild
isolates on the inoculated plants and in their ability to
prevent detrimental effects caused by the T66a challenge
isolate at different temperatures. At warm temperatures, the
effect of the CTV mild isolates on performance of both
Valencia/sour orange and Valencia/macrophylla were negligible.
The decline indexes for the healthy uninoculated control
plants were nearly egual to or greater than those inoculated


BIOGRAPHICAL SKETCH
Mario Alberto Rocha-Pea was borne in Monterrey, Nuevo
Len, Mxico, on July 30, 1950. He received the degree of
Bachelor of Science in microbiology in 1977 from the
Universidad Autnoma de Nuevo Len, and in 1979 received his
Master of Science degree in plant pathology at the Colegio de
Postgraduados, Chapingo, Mxico. From 1979 to 1983 he served
as a professor in the Department of Plant Pathology at the
Colegio Superior de Agricultura Tropical, Tabasco. In 1984
he accepted a position as a research plant pathologist at the
Instituto Nacional de Investigaciones Forestales y
Agropecuarias (INIFAP), in his home state of Nuevo Len, where
he was working until December 1986, before he came to United
States to pursue the degree of Doctor of Philosophy in plant
pathology at the University of Florida.
104


CITRUS TRISTEZA VIRUS: CROSS-PROTECTION, GRAFT
TRANSMISSION, AND SEROLOGY
By
MARIO ALBERTO ROCHA-PEA
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
1990


41
being used as inoculum. All donor host plants were known to
be CTV infected for at least one year before the study was
started.
Grafting procedures and receptor hosts. Rectangular leaf
and bark pieces of about 3 X 15 mm were cut from donor hosts
with a sharp knife and inserted under corresponding bark flaps
cut on the stem of one-year-old Madam Vinous sweet orange,
Mexican lime, and grapefruit (C. paradisi Macf.) plants,
herein referred to as receptor hosts. A portion of the
grafted tissue (2-3 mm) was left exposed at the top of bark
flaps to monitor tissue survival at 21 days post-inoculation.
A minimum of five plants of each receptor host were each
inoculated with 4 pieces of either leaf or bark tissue for
every donor host/virus isolate combination tested.
Serological indexing by the double antibody sandwich enzyme-
linked immunosorbent assay (DAS-ELISA) (see below) was carried
out on receptor hosts at three and five months post
inoculation.
Inoculated receptor plants were grown in a commercial
potting mixture (Pro-mix BX) in three liter plastic
containers, and fertilized with a mixture of NPK (20-10-20)
every other week, and given disease and pest management as
described in Chapter 2.
Virus distribution and antigen concentration in host
tissues. Individual Madam Vinous sweet orange and C. excelsa
plants infected with CTV isolates T26 or T66a were used to


Table 3.4 Relative antigen titer of citrus tristeza virus in different tissues of Citrus
excelsa and Madam Vinous sweet orange host plants, as measured by enzyme-linked immunosorbent
assay.
Host tissue
Virus
isolate
Donor host
Bark
Petioles
Midribs
Leaf
blade
T26
Citrus excelsa
0.349-^/a-/
0.137 a
0.086 a
0.044
a
Madam Vinous
0.221 b
0.188 a
0.120 a
0.065
a
T66a
Citrus excelsa
0.336 a
0.238 a
0.133 a
0.030
a
Madam Vinous
0.266 a
0.099 b
0.040 b
0.029
a
/ Mean of optical density (OD4Q5) per 10 mg plant tissue after 120 min of substrate reaction.
There were four replicates per plant and four plants per host/isolate combination.
Control reaction with the same tissue from healthy plants averaged OD4Q5 = 0.001-0.025.
This has not been subtracted from the values above.
! Numbers in the same column per host/isolate combination followed by different letters
are statistically different by Duncan's test (P < 0.05).


59
laborious and time-consuming for large scale indexing, and
both antibodies and large volumes of buffer are used in each
test.
A simple and rapid serological method, known as dot-
immunobinding assay (DIBA) (Hawkes et al. 1982), has been
developed and applied for the detection of several plant
viruses (Hibi and Saito, 1985; Powell, 1987). The principles
of DIBA are similar to those of ELISA, differing mostly in
that the antigen and antibodies are bound to nitrocellulose
membranes instead of polystyrene microtiter plates, and that
the product of the enzyme reaction at the end of the test is
insoluble. The use of nitrocellulose membranes for the
serological detection of plant viruses has become popular
because of the simplicity of equipment required and lower cost
of materials needed. In a preliminary study the DIBA test was
as sensitive as DAS-ELISA to detect CTV in both field trees
and greenhouse grown plants (Rocha-Pea et al. 1990); however,
some differences were found in the reactivity of the
antibodies used, and some nonspecific reactions occurred in
the test. The objective of this research was to adapt DIBA for
diagnosis of CTV using different polyclonal and monoclonal
antibodies and to compare the sensitivity of DIBA with DAS-
ELISA and DAS-indirect ELISA.
Materials and Methods
Antisera used. Polyclonal antisera numbers 1051, 1052,
and 1053 previously prepared in rabbits against undegraded,


91
There were some differences in the virus concentration
in the donor tissues used as inoculum. Bark tissue contained
the highest titer with OD405 values in the range of 0.221 and
0.349 in both Madam Vinous and C. excelsa with both CTV
isolates tested. These values were, in some instances, more
than double those found in petioles and midribs, and at least
triple those found in the leaf blade. While these values were
not statistically significant, some differences were found in
virus titer from different parts of the same plant and from
one plant to another. There were some instances where OD405
values were as low as the healthy controls indicating a
possible absence of the virus in those tissues. This raises
the possibility that occasionally the tissue used for graft
transmission may be virus-free, with a subsequent failure in
the transmission.
The use of leaf and/or bark pieces for graft transmission
of CTV may be advantageous when large numbers of plants are
to be inoculated with limited sources of inoculum, as is the
case for studies of cross-protection. However, in order to
achieve a high level in the rate of transmission, the
efficiency of donor host and the donor/receptor host
combination should be considered. Leaf pieces provide a
better source for inoculation than bark tissue.
A Dot-Immunobindinq Assay for Citrus Tristeza Virus
The dot-immunobinding assay (DIBA) was adapted for
detection of CTV. Some differences were found in the
J


Table 4.1 Relative
polyclonal (1053) and
tristeza virus.
sensitivity
monoclonal
Of DIBA,
(3DF1 and
DAS-ELISA and DAS-
MCA-13) antibodies
indirect
specific
ELISA with
to citrus
CTV T-3 6
isolate'
DIBA
DAS-ELISA1/
DAS-indirect
ELISA1/
1053
3DF1
MCA-13
1053
3DF1
MCA-13
1/10
' +1/
+
+
0.572
+4/
2.628 +
0.892 +
1/20
+
+
+
0.460
+
2.507 +
0.607 +
1/40
+
+
+
0.314
+
1.635 +
0.385 +
1/80
+
+
+
0.185
+
0.853 +
0.252 +
1/160
+
+
+
0.109
+
0.518 +
0.158 +
1/320
+
+/-
+/-
0.070
-
0.190 +
0.094 -
1/640
-
-
-
0.024
-
0.075 -
0.054 -
1/1280
-
-
-
0.022
-
0.032 -
0.036 -
C. excelsa
-

-
0.004
-
0.033 -
0.013 -
The IgG of polyclonal antibody no. 1053 was used to coat the plates for both DAS-
ELISA and DAS-indirect ELISA. For DAS-ELISA the no. 1053 IgG conjugate was used as
second antibody. For DAS-indirect ELISA the unlabeled 3DF1 and MCA-13 were the
intermediate antibodies followed by the goat anti-mouse IgG conjugate.
1/


62
solution. The substrate solution was prepared as follows: 10
mg of nitro blue tetrazolium (NBT) (Sigma) were dissolved in
30 ml of TBS substrate buffer (0.1 M Tris, 0.1 M NaCl, 0.005
M MgCl2, pH 9.5); then 5 mg of 5-bromo-4-chloro-3-indoyl
phosphate (BCIP) (Sigma) were added and dissolved in the
solution. The color reaction was stopped by transferring the
membranes to distilled water.
DAS-ELISA.- The double antibody sandwich enzyme-linked
immunosorbent assay (DAS-ELISA) (Bar-Joseph et al. 1979b,
1980) was conducted using polyclonal IgG No. 1053 and
polystyrene Immulon II microtiter plates (Dynatech
Laboratories). Unless otherwise stated, 200 microliters were
used per well and three washings with PBS-Tween (phosphate
buffered saline = 8 mM Na2HP04, 14 mM KH2P04, 15 mM NaCl, pH
7.4, + 0.05% Tween 20) were performed between steps.
Microtiter plates were coated using 3.0 ng/ml of purified CTV
specific IgG in carbonate buffer (0.015 M NaHC03, 0.03 M NaC03,
pH 9.6) and incubated for 6 hr at 37C. Antigen samples were
added to the wells and incubated for 18 hr at 5C. CTV
specific IgG conjugated with alkaline phosphatase at a
dilution of 1:1,000 in conjugate buffer {PBS-Tween + 2%
polyvinyl pyrrolidone (=PVP 40,000 MW) (w/v) + 0.2% bovine
serum albumin (w/v)} was incubated for 4 hr at 37C. The
reaction with p-nitrophenyl phosphate (Sigma) (1.0 mg/ml in
10% triethanolamine, pH 9.8) was quantified after 60 min at


18
serological evaluation was made five months after inoculation
(Table 2.2).
The OD405 values were generally higher for the test plants
grown at cooler temperatures. The Valencia/sour orange plants
inoculated with mild isolates and unchallenged with T66a gave
OD405 values between 0.138 and 0.405 when analyzed with PCA.
The corresponding healthy plants averaged 0.039. The same
treatments, including healthy controls, gave values in the
range of 0.016-0.078 with MCA-13. Treatments pre-inoculated
with mild isolates and challenged with T66a gave OD405 values
in the range of 0.174-0.548 with PCA and 0.047-0.477 the MCA-
13. The control plants uninoculated with mild isolates but
challenged with T66a gave OD405 values of 0.292 with PCA and
0.283 with MCA-13 (Table 2.3).
Also at cooler temperatures the Valencia/macrophylla
plants inoculated with mild isolates and unchallenged with
T66a, gave OD405 values between 0.3 35 and 0.361 with PCA. The
value for the corresponding uninoculated healthy control
plants was 0.036. The same treatments, including healthy
controls, gave values in the range of 0.013-0.075 when
analyzed with MCA-13. Treatments pre-inoculated with mild
isolates and challenged with T66a gave values in the range of
0.406-0.456 with PCA and 0.166-0.320 with MCA-13. The control
plants uninoculated with mild isolates but challenged with
T66a gave values of 0.432 with PCA and 0.421 with MCA-13
(Table 2.4). The Tila and T30 isolates were not evaluated for


CITRUS TRISTEZA VIRUS: CROSS-PROTECTION, GRAFT
TRANSMISSION, AND SEROLOGY
By
MARIO ALBERTO ROCHA-PEA
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
1990

To my beloved wife Alicia, and to my adored son Eric,
whom are the world of my life.
To my parents Herminio and Dolores, and my brothers and
sisters.

ACKNOWLEDGEMENTS
I want to express my sincere gratitude and acknowledgment
to the following persons, institutions, and organizations, who
supported my graduate studies at the University of Florida.
The financial support from Consejo Nacional de Ciencia
y Tecnologa (CONACyT), as well as the support and leave of
absence from Instituto Nacional de Investigaciones Forestales
y Agropecuarias (INIFAP), both institutions from Mxico, is
greatly appreciated.
To my co-major professors, Drs. R.F. Lee and C.L.
Niblett, for their constant encouragement, guidance and
support throughout the course of the thesis work, and for
their personal interest in my academic preparation. To them
and the rest of the supervisory committee, Drs. S.M. Garnsey,
D.E. Purcifull, and R.K. Yokomi, for valuable suggestions in
revision of the manuscript.
The financial support from the Florida High Technology
and Industrial Council, and the Florida Citrus Production
Managers' Association, to carry out some parts of the thesis
work also is acknowledged.
I thank Drs. S.M. Garnsey and T.A. Permar, USDA Orlando,
and Drs. P. Moreno and M. Cambra, IVIC Valencia (Spain), for
supplying the MCA-13 and 3DF1 monoclonal antibodies,
respectively.
ii

The technical assistance and friendship of N. Berger, S.
Marquardt, T. Nguyen, S. Jackson, and J. Zellers, as well as
the help of T. Zito in the photographic work, and M. Ahnger
with the statistical analysis, all at the Citrus Research and
Education Center, at Lake Alfred, is greatly appreciated.
The warm friendship I found in my fellow students, lab
technicians, administrative staff, and most faculty members
at the Plant Pathology Department, Gainesville, and at the
Citrus Research and Education Center, Lake Alfred, was indeed
encouraging and will be unforgettable.
Finally, I want to thank my wife Alicia for her constant
encouragement and patience to endure the hardship of my
pursuit of this graduate degree.
iii

TABLE OF CONTENTS
Page
ACKNOWLEDGMENTS
LIST OF FIGURES vi
LIST OF TABLES ix
ABSTRACT xi
CHAPTER
1. INTRODUCTION 1
2. EVALUATION OF THE PROTECTING EFFECTS OF SOME
FLORIDA ISOLATES OF CITRUS TRISTEZA VIRUS AGAINST
THE DEVELOPMENT OF THE DECLINE SYNDROME 6
Introduction 6
Materials and Methods 9
Virus isolates and donor hosts 9
Inoculation of CTV isolates and receptor hosts. 10
Serological tests 11
Detrimental effects of mild isolates and
evaluation of cross-protection 14
Results 14
Antigen titers of mild and severe CTV
isolates with polyclonal and MCA-13
monoclonal antibodies 14
Detrimental effect of mild isolates and
evaluation of cross-protection 21
Discussion 27
3. EFFECTIVENESS OF DIFFERENT CITRUS SPECIES AS
DONOR HOSTS FOR GRAFT TRANSMISSION OF CITRUS
TRISTEZA VIRUS 38
Introduction 38
Materials and Methods 40
Virus isolates and donor hosts 40
Grafting procedures and receptor hosts 41
Virus distribution and antigen
concentration in host tissues 41
Purification of CTV 42
iv

Serological tests 42
Results 43
Graft transmission of citrus tristeza
virus isolates 43
Virus distribution and antigen
concentration in host tissues 47
Discussion 52
4. DEVELOPMENT OF A DOT-IMMUNOBINDING ASSAY FOR
CITRUS TRISTEZA VIRUS 57
Introduction 57
Materials and Methods 59
Antisera used 59
Sample preparation 60
Dot-immunobinding assay 61
DAS-ELISA 62
DAS-indirect ELISA 63
Evaluation 63
Evaluation of different buffers for
sample extraction 64
Results 65
Development of the dot-immunobinding assay 65
Evaluation 69
Evaluation of different buffers for
sample extraction 77
Discussion 77
5. SUMMARY AND CONCLUSIONS 86
LITERATURE CITED 94
BIOGRAPHICAL SKETCH 104
v

LIST OF FIGURES
Page
Figure 2.1 Plot of purified citrus tristeza
virus (CTV) against optical density.
Bark of healthy Citrus excelsa
(0.5 g) tissue was ground in 5.0 ml
of phosphate buffered saline, pH 7.6,
+ 0.05% Tween + 2% polyvinyl pyrrolidone
and mixed with purified CTV T26 isolate
to give the desired optical density
at 260 nm (OD260) DAS-ELISA was
performed as described in materials
and methods. An extinction
coefficient of 2.0 was assumed
(Gonsalves et al. 1978) to estimate
the relative virus concentration 22
Figure 3.1 Plot of purified citrus tristeza
virus (CTV) against optical density.
Bark of healthy Citrus excelsa
(0.25 g) tissue was ground in 5.0 ml
of phosphate buffered saline, pH 7.6,
+ 0.05% Tween + 2% polyvinyl pyrrolidone
and mixed with purified CTV T26 isolate
to give the desired optical density
at 260 nm (OD260) DAS-ELISA was
performed as described in materials
and methods. An extinction
coefficient of 2.0 was assumed
(Gonsalves et al. 1978) to estimate
the relative virus concentration 51
Figure 4.1 Effect of different blocking solutions
on the reaction of polyclonal antibodies
no. 1053 in dot-immunobinding assay with
citrus tristeza virus (CTV) isolates:
A. TBS alone; B. 3% bovine serum albumin
(BSA); C. 3% gelatin; D. 0.5% non-fat dry
milk; E. 5% Triton X-100. Number 1,
CTV T66a; 2, CTV T26; 3, buffer extract
from healthy sweet orange plants;
4, TBS-Tween. Reaction conditions are
described in materials and methods 66
Figure 4.2 Reaction of polyclonal and monoclonal
antibodies in dot-immunobinding assay
with citrus tristeza virus (CTV) isolates.
A. Polyclonal 1053 (1.0 ig/ml) not cross-
vi

absorbed with buffer extract of healthy
plant. B. Polyclonal 1053 (1.0 /g/ml)
incubated with 1:200 (v/v) buffer extract
from healthy plants for 2 hr at 37C
prior to use. C and D, monoclonal 3DF1
(1.0 nq/ml) and MCA-13 (1:5,000 dilution)
antibodies not cross-absorbed with healthy
extract. Number 1, CTV T66a; 2, CTV T26;
3, buffer extract from healthy sweet orange
plants; 4, TBS-Tween. Reaction conditions
are described in materials and methods 68
Figure 4.3 Relative sensitivity level of different
polyclonal and monoclonal antibodies
specific to citrus tristeza virus (CTV)
in dot-immunobinding assay. Rows A, B and C,
polyclonal antibodies nos. 1051, 1052, and
1053, respectively. Rows D and E,
monoclonal antibodies 3DF1 and MCA-13,
respectively. Extract of Citrus excelsa
greenhouse grown plants infected with
CTV T36 isolate was prepared in
TBS-Tween 1:10 (w/v) and two fold
diluted with extract of healthy plants.
Reaction conditions are described in
materials and methods 70
Figure 4.4 Reaction of polyclonal antibodies no.
1053 and 3DF1 and MCA-13 monoclonal
antibodies in dot-immunobinding assay
with twelve selected citrus tristeza
virus (CTV) isolates. Row A: 1 = Tila;
2 = T26; 3 = T30; 4 = T50a; 5 = T55a;
6 = T3; 7 = T4; 8 = T36. Row B. 1 = T62a;
2 = T65a; 3 = T66a; 4 = T67a. Also in Row B
are extracts of healthy plants: 5 = Citrus
excelsa; 6 = Madam Vinous sweet orange;
7 = Duncan grapefruit; 8 = Mexican lime.... 74
Figure 4.5 Evaluation of different extraction
buffers on the sensitivity of DIBA with
citrus tristeza virus (CTV) isolates T26,
T62a and T66a. TBS = Tris buffered saline,
TBST = TBS containing 0.05% Tween, PBS =
phosphate buffered saline, PBST = PBS +
0.05% Tween, Carb = carbonate buffer, CarbT
= Carb + 0.05% Tween, HC = healthy control.
The IgG of polyclonal antibodies No. 1053
or monoclonal antibody 3DF1 were used as
primary antibodies followed by the goat anti
rabbit and anti- mouse IgG conjugates,
respectively. Samples were 2 ;xl of buffer
vii

extracts of greenhouse-grown Madam Vinous
sweet orange plants infected with the CTV
isolates ground at a 1:10 dilution 78
Figure 4.6 Evaluation of different extraction
buffers on the sensitivity of DAS-ELISA
and DAS-indirect ELISA with citrus tristeza
virus (CTV) isolates T26, T62a and T66a.
TBS = Tris buffered saline, TBST = TBS+Tween
(0.05%), PBS = phosphate buffered saline,
PBST = PBS+Tween (0.05%), Carb = carbonate
buffer, CarbT = Carb+Tween (0.05%),
HC = healthy control. The IgG of
polyclonal antibodies No. 1053 was used
to coat the plates for both DAS-ELISA and
DAS-indirect ELISA. For DAS-ELISA the No.
1053 IgG conjugate was used as second
antibody. For DAS-indirect ELISA the
unlabeled 3DF1 monoclonal antibody was the
intermediate antibody followed by the goat
anti-mouse IgG conjugate. Samples were
200 nl of buffer extracts of
greenhouse-grown Madam Vinous sweet orange
plants infected with the CTV isolates
ground at a 1:10 dilution 79
viii

LIST OF TABLES
Table 2.
Table 2.
Table 2.
Table 2.
Table 2.
Table 2.
Page
1 Relative antigen titer of citrus
tristeza virus (CTV) mild isolates
in plants unchallenged and
challenged by the T66a severe
isolate: I. Warm temperature,
Valencia/sour orange 16
2 Relative antigen titer of citrus
tristeza virus (CTV) mild isolates
in plants unchallenged and
challenged by the T66a severe
isolate: II. Warm temperature,
Valencia/macrophylla 17
3 Relative antigen titer of citrus
tristeza virus (CTV) mild isolates
in plants unchallenged and
challenged by the T66a severe
isolate: III. Cool temperature,
Valencia/sour orange 19
4 Relative antigen titer of citrus
tristeza virus (CTV) mild isolates
in plants unchallenged and
challenged by the T66a severe
isolate: IV. Cool temperature,
Valencia/macrophylla 20
5 Effect of citrus tristeza virus (CTV)
mild isolates on the development of the
CTV decline syndrome in plants
unchallenged and challenged by the
T66a severe isolate: I. Warm temperature,
Valencia/sour orange 23
6 Effect of citrus tristeza virus (CTV)
mild isolates on the development of the
CTV decline syndrome in plants
unchallenged and challenged by the
T66a severe isolate: II. Warm temperature,
Valencia/macrophylla 25
ix

Table 2.7
Table 2
Table 3
Table 3
Table 3
Table 3
Table 4
Table 4
Effect of citrus tristeza virus (CTV)
mild isolates on the development of the
CTV decline syndrome in plants
unchallenged and challenged by the
T66a severe isolate: III. Cool temperature,
Valencia/sour orange 26
8 Effect of citrus tristeza virus (CTV)
mild isolates on the development of the
CTV decline syndrome in plants
unchallenged and challenged by the
T66a severe isolate: IV. Cool temperature,
Valencia/macropylla 28
1 Transmission of citrus tristeza virus by
graft inoculation between selected
citrus hosts: I. Efficiency of leaf and
bark pieces as inoculum 45
2 Transmission of citrus tristeza virus by
graft inoculation between selected
citrus hosts: II. Overall rate of
transmission 46
3 Transmission of citrus tristeza virus by
graft inoculation between selected
citrus hosts: III. Effect of virus
isolates 48
4 Relative antigen titer of citrus
tristeza virus in different tissues of
Citrus excelsa and Madam Vinous sweet
orange host plants, as measured by enzyme-
linked immunosorbent assay 50
1 Relative sensitivity of DIBA, DAS-ELISA and
DAS-indirect with polyclonal (1053) and
monoclonal (3DF1 and MCA-13) antibodies
specific to citrus tristeza virus 71
2 Comparison of DIBA, DAS-ELISA and
DAS-indirect ELISA for the detection
of citrus tristeza virus (CTV) and
relative reactivity of polyclonal
(1053) and monoclonal (3DF1) (MCA-13)
antibodies with CTV isolates 75
x

Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
CITRUS TRISTEZA VIRUS: CROSS-PROTECTION, GRAFT
TRANSMISSION, AND SEROLOGY
By
MARIO ALBERTO ROCHA-PEA
DECEMBER 1990
Chairperson: Dr. Richard F. Lee
Major Department: Plant Pathology
The objectives of this research were i) to evaluate some
citrus tristeza virus (CTV) mild isolates under greenhouse
conditions for cross protecting ability against the decline
syndrome, and ii) to develop methods for detection of severe
CTV challenge isolates in mixed infections in cross-protection
experiments. Valencia sweet orange plants budded on sour
orange rootstock were graft-inoculated by leaf pieces using
any of four different mild CTV isolates and subsequently
graft-challenged with a severe CTV isolate. Treatments were
evaluated at temperature regimens of 21-38C and 21-33C.
Plants pre-inoculated with mild isolates when challenged with
the severe isolate gave relatively lower ELISA values as
compared to the unprotected, challenged control plants. The
MCA-13 monoclonal antibody provided a rapid method to detect
xi

the severe isolate in mixed infections. CTV-induced decline
(CTV-ID) occurred irregularly within the first 10 months after
challenge inoculation at both temperature regimens. The
preliminary evaluation of the cross-protecting ability of mild
isolates against the CTV-ID in plants on sour orange rootstock
can be accomplished under greenhouse conditions in a
relatively short time of 18-24 months.
Differences were found in the effectiveness of certain
tissues and/or hosts for graft-transmission of CTV. Leaf-
piece grafts transmitted CTV at a 90% rate vs a 75% rate using
bark pieces. Madam Vinous sweet orange was the most efficient
donor host giving 90% transmission to three receptor hosts,
followed by Mexican lime at 85%, and Citrus excelsa at 72%.
The dot-immunobinding assay (DIBA) was adapted for CTV
diagnosis by using several polyclonal and monoclonal
antibodies specific for CTV. The DIBA was as sensitive as
DAS-ELISA and DAS-indirect ELISA for CTV detection and
provides a reliable alternative for diagnosis of CTV.
xii

CHAPTER 1
INTRODUCTION
Citrus tristeza virus (CTV) is distributed in citrus
producing areas worldwide and is the most economically
important viral disease of citrus (Bar-Joseph et al. 1979a,
1981, 1989). The virus infects nearly all species, varieties,
and intergeneric hybrids of citrus, and some citrus relatives
(Bar-Joseph et al. 1979a; Garnsey and Lee, 1988; Muller and
Garnsey, 1984). However, the most destructive damage is the
induced decline in scions grafted on sour orange (Citrus
aurantium L.) rootstock. Some CTV isolates cause stem pitting
and loss of plant vigor on some orange and grapefruit scions
regardless of the rootstock (Bar-Joseph et al. 1979a, 1981,
1989) .
Citrus tristeza virus is a phloem-limited, flexuous
closterovirus approximately 2,000 x 11 nm in size, transmitted
by aphids in a semi-persistent manner (Bar-Joseph et al.
1979a; Lister and Bar-Joseph, 1981). A single stranded
positive sense RNA of 5.4-6.5 x 106 daltons has been isolated
from purified virus preparations (Bar-Joseph et al. 1985).
Several coat proteins of about Mr 28,000 (Guerri et al. 1990),
1

2
23,000 and 21,000 (Lee et al. 1988b), respectively, have been
associated with CTV virions. The virus is readily transmitted
by budding and grafting (Bennett and Costa, 1949; Bar-Joseph
and Lee, 1990; Bar-Joseph et al. 1979a). Mechanical
transmission has been accomplished by slash inoculation of
partially purified virus preparations into the stem of hosts
such as citron (Citrus medica L.) and Mexican lime {C.
aurantifolia (Christm.) Swing.} (Garnsey and Muller, 1988;
Garnsey et al. 1977; Muller and Garnsey, 1984). Seed
transmission has not been demonstrated (McClean, 1957;
Wallace, 1978).
Citrus tristeza virus occurs naturally with a diversity
of isolates or strains which may differ greatly in their
biological properties, such as symptomatology in different
citrus hosts (Garnsey et al. 1987; McClean, 1974), aphid
transmissibility (Bar-Joseph and Loebenstein, 1973; Bar-Joseph
et al. 1977; Roistacher, 1981; Yokomi and Garnsey, 1987), and
sensitivity to warm temperatures (Ieki and Yamada, 1980;
Roistacher et al. 1974).
Several sensitive and relatively rapid methods have been
developed to diagnose the presence of CTV in infected plants.
These methods include SDS-immunodiffusion procedures (Garnsey
et al. 1979; Bar-Joseph et al. 1980), enzyme-linked
immunosorbent assay (ELISA) (Bar-Joseph et al. 1979b, 1980),
light and electron microscopy (Brlansky, 1987; Brlansky et al.
1984; Garnsey et al. 1980a), and in situ immunofluorescence

3
(Brlansky et al. 1984; Tsuchizaki et al. 1978). Each of these
methods has different advantages, disadvantages, and
sensitivity levels. Therefore, a certain method may be used
for a particular purpose. Some methods, such as ELISA, are
dependable and widely used for indexing purposes (Garnsey et
al. 1981a).
Recently a monoclonal antibody was developed against a
decline-inducing isolate from Florida which, in ELISA, reacted
specifically with several severe CTV isolates from diverse
geographical areas, but not with mild isolates from the same
areas (Permar et al. 1990).
Control of CTV is difficult. In those few areas of the
world where CTV still is not present, quarantine and virus-
free certification programs are maintained to prevent the
introduction of infected budwood sources (Bar-Joseph et al.
1983, 1989). Likewise, in those areas with low CTV incidence,
large scale surveys and suppression measures are carried out
to reduce disease spread to other trees and locations and
prolong the use of sour orange as a rootstock (Bar-Joseph et
al. 1989). Once the disease becomes endemic, two situations
can result: a) CTV-induced decline develops and kills plants
grafted onto sour orange rootstock, whereas tolerant
rootstocks do not decline, and/or b) CTV-stem pitting can
affect sweet orange and/or grapefruit scions regardless of the
rootstock, resulting in a loss of plant vigor and yield. Mild
strain cross protection is the only known control measure

4
which is effective against stem pitting (Bar-Joseph et al.
1989; Garnsey and Lee, 1988; Lee et al. 1987a). Genetic
resistance to CTV is not available in commercially acceptable
scions (Bar-Joseph et al. 1989; Garnsey and Lee, 1988). The
application of genetically engineered cross protection,
currently effective in several other crops (Beachy et al.
1987), is an attractive possibility for CTV control in the
future.
CTV has been widespread in Florida for many years and
induced decline has occurred in localized areas (Garnsey and
Jackson, 1975; Norman et al. 1961). However, until recently,
it had not caused major losses because most of the citrus
acreage had been propagated on CTV-tolerant rootstocks and
because of the prevalence of mild CTV isolates which did not
seriously affect trees grafted on sour orange rootstock
(Brlansky et al. 1986; Garnsey et al. 1980b; Lee et al.
1987a) In the last decade the situation in Florida has
changed radically. Sour orange continued to be a very popular
rootstock because of cold tolerance, high fruit quality of the
scion, and its tolerance to citrus blight. The high demand
for plants on sour orange, plus discovery of citrus bacterial
leaf spot in some nurseries (Brlansky, 1988; Garnsey, personal
communication), caused nurserymen to use budwood from source
trees that had not been propagated previously on sour orange.
Many such trees apparently were harboring severe CTV isolates
(Brlansky et al. 1986; Lee et al. 1987a). Severe dwarfing of

5
young trees propagated on sour orange has appeared in many
parts of Florida. Large scale outbreaks of induced decline
also have appeared in southern Florida, an area previously not
affected by CTV. Losses have exceeded 50% in some plantings
(Brlansky et al. 1986).
Management of CTV-induced decline in Florida is
difficult. The effective use of CTV tolerant rootstocks, such
as rough lemon (Citrus iambhiri Lush.), Troyer citrange
{Ppncirus trifoliata (L.) Raf. x C. sinensis (L.) Osb.},
Cleopatra mandarin (C. reshni Hort. ex Tanaka), sweet orange
(C. sinensis) and others (Grant et al. 1961; Wallace, 1978)
is diminished by their susceptibility to other diseases, most
importantly citrus blight, an endemic disease of unknown
etiology which is removing more than 500,000 trees from
production annually (Lee et al. 1988a).
The objectives of this research were: i) To evaluate some
CTV mild isolates under greenhouse conditions for cross
protecting ability against the decline syndrome, and ii) To
develop methods for detection of the severe CTV challenge
isolate in mixed infections.

CHAPTER 2
EVALUATION OF THE PROTECTING EFFECTS OF SOME MILD FLORIDA
ISOLATES OF CITRUS TRISTEZA VIRUS AGAINST THE DEVELOPMENT OF
THE DECLINE SYNDROME
Introduction
Cross protection is a control strategy used to reduce
losses due to plant viral diseases by the use of mild or
attenuated strains of a virus which prevent the effect or
expression of a related, and usually more severe, strain of
the same virus (Fulton, 1986). Cross protection can be useful
when the virus disease is endemic, causes great losses, and
no host genetic resistance is available (Fulton, 1986;
Gonsalves and Garnsey, 1989: Hamilton, 1985; Muller et al.
1982). Cross protection has been used commercially to control
CTV stem pitting isolates on Pera sweet orange in Brazil
(Costa and Muller, 1980; Muller, 1980), and is a part of South
Africa's citrus cultivar improvement program to reduce CTV-
induced stem pitting on grapefruit (DeLange et al. 1980;
Garnsey and Lee, 1988). Relatively little work has been done
to evaluate the potential of cross protection against the CTV-
induced decline (CTV-ID) on sour orange, as most countries
abandon sour orange as a rootstock when CTV-ID isolates become
prevalent. However, experiments conducted in Australia
6

7
(Thornton et al. 1980), United States (Wallace and Drake,
1976; Yokomi et al. 1990), and Japan (Miyakawa, 1987) indicate
that cross protection against CTV-ID isolates may be possible.
The classical approach for selecting potential cross
protecting mild CTV isolates has been empirical, that is, by
collecting a number of CTV isolates from outstanding trees in
groves severely affected by the disease (Balaraman and
Ramakrishnan, 1980; Muller and Costa, 1987). Those isolates
are propagated on several scion-rootstock combinations in
nursery plants and further evaluated over a period of several
years under field conditions (Muller, 1980; Muller and Costa,
1977) This requires the handling and care of large numbers
of plants, extensive field space, and considerable time; the
results are evaluated over a period of 5-12 years. By this
approach, only six mild CTV isolates out of 45 mild isolates
originally selected in Brazil were useful for cross protection
(Costa and Muller, 1980).
Another approach for the selection of mild CTV isolates
for cross protection relies upon host effects for the strain
segregation of field collected severe CTV isolates (Roistacher
et al. 1988). The selection of potential cross protecting CTV
isolates was made from either symptomless or recovered
infected plants after extensive passage of the virus through
a series of different citrus and non-citrus hosts in the
greenhouse (Roistacher et al. 1987, 1988). This approach
provided nine mild or attenuated CTV isolates with outstanding

8
protection against either seedling yellows or stem pitting CTV
isolates from a total of 116 evaluated when challenged with
the original isolates from which they were derived. This
approach required extensive greenhouse work to follow the
segregation of every single isolate, and some attenuated
isolates showed a tendency to revert back to their severe
forms.
Recently, the greenhouse evaluation of CTV-stem pitting
isolates as protecting agents against the CTV-ID on sweet/sour
orange combinations was reported (Miyakawa, 1987). However,
CTV-stem pitting isolates as protecting agents offer limited
possibilities of commercial use in Florida, particularly
because stem pitting isolates still are not present and
susceptible species or cultivars such as grapefruit are grown
extensively.
Citrus tristeza virus causing decline on trees on sour
orange rootstock is endemic in Florida. The effective use of
CTV-tolerant rootstocks is diminished by their susceptibility
to citrus blight. In addition, the increased popularity of
sour orange despite tristeza, along with the natural
prevalence of mild CTV isolates, provides the opportunity to
evaluate cross protection as an alternative control strategy
for CTV-ID. However, there are several considerations: i)
The threat of recurrent freezes makes it difficult to reliably
evaluate the cross protection potential of mild isolates under
field conditions, especially in the Ridge area; ii) There is

9
a need for a method to more rapidly select and evaluate
potential cross protecting mild isolates which is adaptable
for screening large numbers of isolates; and iii) There is
a need to differentiate mild and severe CTV isolates in a
mixed infection in the same plant to aid in the evaluation and
to better understand the mechanism by which cross protection
functions.
The objectives of this research were to evaluate
naturally occurring Florida mild CTV isolates for cross
protecting ability and to develop a methodology to detect the
presence of protecting and challenge CTV isolates in mixed
infections. The effect of temperature on the cross protecting
ability of CTV mild isolates also was studied.
Materials and Methods
Virus isolates and donor hosts. Five naturally occurring
Florida CTV isolates collected from field grown sweet orange
or grapefruit trees grafted on sour orange were used in these
experiments after transmission by Aphis qossypii Glover. The
Tila, T26, T30 and T55a isolates produce very mild symptoms
and little or no stunting on Mexican lime seedlings {Citrus
aurantifolia (Christm.) Swingle}, no seedling yellows on
Eureka lemon {C. limn (L.) Burm.} or sour orange (C.
aurantium L.) and are symptomless in sweet orange {C. sinensis
(L.) Osb.} and sweet/sour orange combinations (Garnsey et al.
1987; Lee, 1984; Yokomi and Garnsey, 1987; Yokomi et al.
1987) The T66a challenge isolate causes strong vein

10
clearing, stunting and stem pitting in Mexican lime seedlings
and severe decline on sweet/sour orange combinations (Garnsey
et al. 1987; Yokomi et al. 1987). The CTV isolates were
propagated in either C. excelsa Wester, Madam Vinous sweet
orange or Mexican lime plants and maintained in a greenhouse
with mean minimum and maximum temperatures of 21 and 38C,
respectively. Inoculum source tissue from donor hosts was
evaluated by serological indexing (see below) to confirm the
presence of CTV before being used as inoculum.
Inoculation of CTV isolates and receptor hosts. One-
year-old Valencia sweet orange plants budded on either sour
orange or C. macrophvlla Wester rootstocks, were graft
inoculated in the stem with each of the CTV mild isolates
using three leaf pieces or blind buds per plant (Garnsey and
Whidden, 1970; Garnsey et al. 1987). Inoculum tissue was
sealed firmly into the receptor stems with plastic grafting
tape. Three weeks later the grafting tape was removed and
the plants were evaluated for survival of grafted tissue and
reinoculated if the grafted tissue had not survived. After
verifying by serological indexing that infection by mild
isolates had taken place, a minimum of four inoculum pieces
of T66a infected tissue were used to challenge the test
plants, and they were reinoculated if at least two inoculum
pieces were not alive 21 days post-challenge. Surviving
inoculum tissue was left in place for the duration of the
experiment. Inoculated receptor plants were grown in a

11
commercial potting mixture (Pro-mix BX) in five liter plastic
containers, and fertilized with a mixture of NPK (20-10-20)
every other week. Pest and disease management included the
application of 0.300 g active ingredient (a.i.)/plant of
aldicarb and 0.86 g a.i /L soil drench of ridomil twice a
year. The experiment was conducted in a greenhouse with mean
minimum and maximum temperatures of 21 and 38C, respectively.
At least 15 plants were inoculated for each CTV isolate/scion/
rootstock combination.
A second set of one-year old Valencia sweet orange plants
budded on sour orange and C. macrophvlla rootstocks were
inoculated with each of the CTV mild isolates and challenged
with the T66a severe isolate as previously described. These
plants were placed in a greenhouse with controlled mean
minimum and maximum temperatures of 21 and 33C, respectively,
to evaluate the effect of temperature on the cross protecting
ability of the CTV isolates. At least 10 plants were
inoculated per CTV isolate/scion/rootstock combination.
Fertilization and plant pest and disease management was as
above.
Serological tests. CTV infection and relative antigen
titer of inoculated plants were determined throughout the
study by the double antibody sandwich enzyme-linked
immunosorbent assay (DAS-ELISA) (Bar-Joseph et al. 1979b,
1980), using polyclonal antiserum No. 1053 prepared against
whole, unfixed CTV isolate T26 (R.F. Lee, unpublished). The

12
severe CTV isolate T66a was detected in the challenged plants
by DAS-indirect ELISA using the MCA-13 strain specific
monoclonal antibody which reacts strongly against most severe
CTV isolates (Permar et al. 1990). For efficiency the
serological tests were performed on both experiments at the
same time.
Routinely, 0.5 g of bark, petioles and midribs of new,
fully expanded tissue were finely chopped with a razor blade
and ground, using a Tekmar Tissumizer, in 5 ml of phosphate
buffered saline (PBS)-Tween + polyvinyl pyrrolidone {PBS = 8
mM Na2HP04, 14 mM KH2P04, 15 mM NaCl, pH 7.4, (+ 0.1 % Tween
20 + 2% polyvinyl pyrrolidone (PVP-40 Sigma)}. Unless stated
otherwise, 200 microliters samples were used per well of the
microtiter plates and three washings with PBS-Tween (phosphate
buffered saline + 0.1 % Tween 20) were performed between
steps. The immunoglobulins (IgG) present in the whole CTV
antiserum were purified by the Protein A-Sepharose affinity
method (Miller & Stone, 1978). A portion of purified
immunoglobulins were conjugated to alkaline phosphatase by the
glutaraldehyde method (Clark et al. 1986). Polystyrene
Immulon II microtiter plates (Dynatech Laboratories) were
coated with 2.0 /g/ml of purified IgG in carbonate buffer
(0.015 M NaHC03, 0.03 M NaC03, pH 9.6) and incubated for 6 hr
at 37C. Antigen samples were added to the wells and
incubated for 18 hr at 5C. Enzyme conjugate was used at a
dilution of 1:1,000 in conjugate buffer (PBS-Tween + 2%

13
polyvinyl pyrrolidone + 0.2% bovine serum albumin) and
incubated for 6 hr at 37C. The reaction with one mg/ml of p-
nitrophenyl phosphate (Sigma) in 10% triethanolamine, pH 9.8,
was measured after 120 min at 405 nm (OD405) with a
Labinstruments model EAR 400 AT ELISA plate
spectrophotometer. Samples were considered positive when OD405
values were higher than 0.100 or three times the mean of
healthy controls, whichever was greater.
For DAS-Indirect ELISA, the microtiter plates were first
coated with IgG from antiserum No. 1053. Antigen samples were
added as described for DAS-ELISA. The MCA-13 strain specific
monoclonal antibody (hereafter MCA-13), as ascites fluid, was
added at a dilution of 1:5,000 (v/v) in conjugate buffer and
incubated 4 hr at 37C. After washing, goat anti-mouse IgG
labeled with alkaline phosphatase (Promega) at a dilution of
1:7,500 (v/v) in conjugate buffer was added and incubated for
2 hr at 37C. The enzyme reaction was carried out as for DAS-
ELISA.
For all serological tests, two replications were used
per sample. Positive controls included four mild isolates
(Tila, T26, T30, and T55a) and one severe (T66a) CTV isolate.
Negative controls included extraction buffer, and similar
buffer extracts from healthy C. excelsa and Valencia sweet
orange plants. A standard curve prepared with purified CTV
T26 isolate diluted to OD260 values of 0.04, 0.02, 0.01, 0.005,
0.0025, 0.0012, 0.0006, and 0.0003 diluted in buffer extract

14
of healthy C. excelsa was used to estimate the relative
antigen concentration of test samples.
Detrimental effects of mild isolates and evaluation of
cross protection. To evaluate the effect of each CTV isolate
on the inoculated plants and the protecting ability of the
mild isolates against the T66a challenge isolate, evaluations
were made at five and ten months after the challenge
inoculation with the T66a isolate. Phloem necrosis was
evaluated by cutting a bark flap at the bud union and the
plant tissue was examined with a hand lens for browning. A
decline index was assigned for each plant. The parameters
scored were stem diameter, plant growth, and foliage symptoms
for decline. Each parameter was visually rated from 0
(minimum) to 3 (maximum), for a maximum cumulative score of
9 for each plant. A high decline index sometimes was
accompanied by plant death. The decline index for each
treatment was the average of the cumulative scores for all
plants in that treatment.
Results
Antigen titers of mild and severe CTV isolates with
polyclonal and MCA-13 monoclonal antibodies. The antigen
titers expressed as optical density (OD405) values for the
different temperature and host treatments, measured by DAS-
ELISA with polyclonal antibodies (PCA) and DAS-indirect ELISA
with MCA-13, are summarized in Tables 2.1, 2.2, 2.3, and 2.4.
At warm temperatures, Valencia/sour orange plants inoculated

15
with mild isolates but unchallenged with T66a gave OD405 values
between 0.091 and 0.145 when analyzed with PCA. The
corresponding uninoculated healthy control plants averaged
0.039. The same treatments, including the healthy controls
gave values in the range of 0.011-0.025 when analyzed by DAS-
indirect ELISA with MCA-13. Treatments pre-inoculated with
mild isolates and further challenged with T66a gave OD405
values in the range of 0.130-0.217 with PCA and 0.145-0.189
with MCA-13. The control plants uninoculated with mild
isolates but challenged with T66a gave values of 0.174 with
PCA and 0.214 with MCA-13 (Table 2.1).
Also at warm temperatures, the Valencia/macrophylla
plants inoculated with the mild isolates and unchallenged with
T66a, gave OD405 values between 0.092 and 0.164 with PCA. The
value for the corresponding uninoculated healthy control
plants was 0.040. The same treatments, including the healthy
controls gave values in the range of 0.018-0.037 when analyzed
with MCA-13. Treatments pre-inoculated with mild isolates and
challenged with the T66a gave values in the range of 0.185-
0.311 with PCA and 0.104-0.305 with MCA-13. The control
plants uninoculated with mild isolates but challenged with
T66a gave values of 0.194 with PCA and 0.251 with MCA-13
(Table 2.2). The T30 isolate was not evaluated in this
portion of the experiment because not enough plants were
available. The plants pre-inoculated with the Tila isolate
were not protected and declined and died before the

16
Table 2.1 Relative antigen titer of citrus tristeza virus
(CTV) mild isolates in plants unchallenged and challenged by
the T66a severe isolate: I. Warm temperature, Valencia/sour
orange.
Unchallenged17 Challenged1727
CTV OD,0_
isolate
Polyclonal
MCA-
13
Polyclonal
MCA-:
13
Tila
0.101-7afc£7
0.024
a
0.217
a
0.175
a
T26
0.123
ab
0.025
a
0.130
a
0.145
a
T3 0
0.091
ab
0.018
ab
0.212
a
0.189
a
T55a
0.145
a
0.014
b
0.194
a
0.181
a
Healthy
0.039
b
0.011
b
0.174
a
0.214
a
or control
plants uninoculated
with mild isolate
One-year-old plants were graft inoculated with leaf
pieces under bark flaps on the stem from donor plants
infected with the indicated mild CTV isolates.
27 After verifying virus infection with mild isolates by
DAS-ELISA with polyclonal antibodies, the challenged
plants were graft inoculated similarly with the T66a
severe isolate.
-7 Optical density at 405 nm (OD405) was measured after 120 min
of reaction.
-7 Serological detection was carried out by DAS-ELISA with
polyclonal antisera and DAS-indirect ELISA with MCA-13
severe strain specific monoclonal antibody six months
after inoculation. Value is the average of duplicate
assays of the corresponding plants in Table 2.5.
-7 Numbers in the same column followed by different letters
are statistically different by Duncan's test (P < 0.05).

17
Table 2.2 Relative antigen titer of citrus tristeza virus
(CTV) mild isolates in plants unchallenged and challenged by
the T66a severe isolate: II. Warm temperature,
Valencia/macrophylla.
Unchallenged17
Challenged
1/2/
CTV
OD*7
OD',05
isolate
Polyclonal
MCA-13
Polyclonal
MCA-13
Tila
0.092-7a-7
0.026 a
_6/
-
T26
0.132 a
0.030 a
0.185 a
0.104 a
T30
NE-7
NE
NE
NE
T55a
0.164 a
0.037 a
0.311 a
0.305 a
Healthy
0.040 a
0.018 a
0.194 a
0.251 a
or control
plants uninoculated
with mild isolate
17 One-year-old plants were graft inoculated with leaf
pieces under bark flaps on the stem from donor plants
infected with the indicated mild CTV isolates.
-7 After verifying virus infection with mild isolates by
DAS-ELISA with polyclonal antibodies, the challenged
plants were graft inoculated similarly with the T66a
severe isolate.
-7 Optical density at 405 nm (OD405) was measured after 120 min
of reaction.
Serological detection was carried out by DAS-ELISA with
polyclonal antisera and DAS-indirect ELISA with MCA-13
severe strain specific monoclonal antibody six months
after inoculation. Value is the average of duplicate
assays of the corresponding plants in Table 2.6.
-7 Numbers in the same column followed by different letters
are statistically different by Duncan's test (P < 0.05).
-1 = severely diseased plants died before serological
evaluation.
NE = treatment not evaluated.

18
serological evaluation was made five months after inoculation
(Table 2.2).
The OD405 values were generally higher for the test plants
grown at cooler temperatures. The Valencia/sour orange plants
inoculated with mild isolates and unchallenged with T66a gave
OD405 values between 0.138 and 0.405 when analyzed with PCA.
The corresponding healthy plants averaged 0.039. The same
treatments, including healthy controls, gave values in the
range of 0.016-0.078 with MCA-13. Treatments pre-inoculated
with mild isolates and challenged with T66a gave OD405 values
in the range of 0.174-0.548 with PCA and 0.047-0.477 the MCA-
13. The control plants uninoculated with mild isolates but
challenged with T66a gave OD405 values of 0.292 with PCA and
0.283 with MCA-13 (Table 2.3).
Also at cooler temperatures the Valencia/macrophylla
plants inoculated with mild isolates and unchallenged with
T66a, gave OD405 values between 0.3 35 and 0.361 with PCA. The
value for the corresponding uninoculated healthy control
plants was 0.036. The same treatments, including healthy
controls, gave values in the range of 0.013-0.075 when
analyzed with MCA-13. Treatments pre-inoculated with mild
isolates and challenged with T66a gave values in the range of
0.406-0.456 with PCA and 0.166-0.320 with MCA-13. The control
plants uninoculated with mild isolates but challenged with
T66a gave values of 0.432 with PCA and 0.421 with MCA-13
(Table 2.4). The Tila and T30 isolates were not evaluated for

19
Table 2.3 Relative antigen titer of citrus tristeza virus
(CTV) mild isolates in plants unchallenged and challenged by
the T66a severe isolate: III. Cool temperature, Valencia/sour
orange.
Unchallenaed1/
Challenaed1/2/
CTV
ODac3/
OD
isolate
Polyclonal
MCA-13
Polyclonal
MCA-13
Tila
0.404^
2
0.060
a
0.548
a
0.477 a
T26
0.405
a
0.078
a
0.195
b
0.055 c
T30
0.138
a
0.054
a
0.174
b
0.047 c
T55a
0.308
a
0.055
a
0.276
b
0.057 c
Healthy
0.039
b
0.016
a
0.292
b
0.283 b
or control
plants uninoculated
with mild isolate
One-year-old plants were graft inoculated with leaf
pieces under bark flaps on the stem from donor plants
infected with the indicated mild CTV isolates.
After verifying virus infection with mild isolates by
DAS-ELISA with polyclonal antibodies, the challenged
plants were graft inoculated similarly with the T66a
severe isolate.
3/ Optical density at 405 nm (OD405) was measured after 120 min
of reaction.
Serological detection was carried out by DAS-ELISA with
polyclonal antisera and DAS-indirect ELISA with MCA-13
severe strain specific monoclonal antibody six months
after inoculation. Value is the average of duplicate
assays of the corresponding plants in Table 2.7.
Numbers in the same column followed by different letters
are statistically different by Duncan's test (P < 0.05).

20
Table 2.4 Relative antigen titer of citrus tristeza virus
(CTV) mild isolates in plants unchallenged and challenged by
the T66a severe isolate: IV. Cool temperature,
Valencia/macrophylla.
Unchallenged17 Challenged1727
CTV OD,0S^
isolate
Polyclonal
MCA-13
Polyclonal
MCA-
13
Tila
NE^7
NE
NE
NE
T26
0.36 l-7a-7
0.049 a
0.406
a
0.166
a
T30
NE
NE
NE
NE
T55a
0.335 a
0.075 a
0.456
a
0.320
a
Healthy
0.036 b
0.013 b
0.432
a
0.421
a
or control
plants uninoculated
with mild isolate
-7 One-year-old plants were graft inoculated with leaf
pieces under bark flaps on the stem from donor plants
infected with the indicated mild CTV isolates.
27 After verifying virus infection with mild isolates by
DAS-ELISA with polyclonal antibodies, the challenged
plants were graft inoculated similarly with the T66a
severe isolate.
-7 Optical density at 405 nm (OD405) was measured after 120 min
of reaction.
-7 NE = treatment not evaluated.
-7 Serological detection was carried out by DAS-ELISA with
polyclonal antisera and DAS-indirect ELISA with MCA-13
severe strain specific monoclonal antibody six months
after inoculation. Value is the average of duplicate
assays of the corresponding plants in Table 2.8.
Numbers in the same column followed by different letters
are statistically different by Duncan's test (P < 0.05).

21
the Valencia/macrophylla combination (Table 2.4) because not
enough plants were available.
From the standard curve prepared with purified T26 (Fig.
2.1), it was estimated that an OD405 value of 0.638 was
approximately equivalent to 20 g/ml of CTV antigen, assuming
an extinction coefficient of 2.0 (Gonsalves et al. 1978).
Therefore, there was an average of 3.1 ng of CTV antigen per
every 100 mg of tissue for each 0.100 OD405 value in the test
samples.
Detrimental effects of mild isolates and evaluation of
cross protection. The protecting effect of mild isolates was
evaluated on the basis of their ability to prevent the
detrimental effects on stem diameter, plant growth and foliage
symptoms caused under greenhouse conditions by the T66a severe
decline isolate. The detrimental effects of mild isolates
alone also were evaluated. The number of plants and the
scores for the decline index established in every treatment
are shown in Tables 2.5, 2.6, 2.7, and 2.8. At warm
temperatures, Valencia/sour orange plants inoculated with mild
isolates and unchallenged with T66a, and the healthy
uninoculated controls, gave overall decline index values in
the range of 0.0 and 1.5. Whereas, the plants pre-inoculated
with mild isolates and challenged with T66a showed higher
decline index values of 6.3, 2.6, 4.2, and 5.5 for the Tila,
T26, T30, and T55a mild isolates, respectively. The control
plants uninoculated with mild isolates but challenged with

22
E
c
ID
O
*->
C\3
>.
*->
CO
c
Q)
Q
o
4
Q.
o
Purified virus (Optical Density 260 nm)
Figure 2.1 Plot of purified citrus tristeza virus (CTV)
against optical density. Bark of healthy Citrus excelsa (0.5
g) tissue was ground in 5.0 ml of phosphate buffered saline,
pH 7.6, + 0.05% Tween + 2% polyvinyl pyrrolidone and mixed
with purified CTV T26 isolate to give the desired optical
density at 260 nm (OD260) DAS-ELISA was performed as described
in materials and methods. An extinction coefficient of 2.0
was assumed (Gonsalves et al. 1978) to estimate the relative
virus concentration.
OJb /s

Table 2.5 Effect of citrus tristeza virus (CTV) mild isolates on the development of
the CTV decline syndrome in plants unchallenged and challenged by the T66a severe
isolate: I. Warm temperature, Valencia/sour orange.
Unchallenged
1/
Challenged
1/2/
CTV
No. of
Decline
index'
No. of
Decline
No. of dead
isolate
plants
plants
index
plants 10 months
after challenge
Tila
4
0.0
8
6.3
3/8
T26
4
1.5
5
2.6
0/5
T30
5
1.0
4
4.2
1/4
T55a
5
0.0
7
5.5
2/7
Healthy
or control
5
1.4
6
4.0
1/6
plants uninoculated
with mild isolate
/ One year
old plants were graft inoculated
with leaf
pieces under bark flaps on the
stem from donor
plants infected
with the indicated
mild CTV isolates.
/ After verifying
virus infection
with mild
isolates
by DAS-ELISA with polyclonal
to
u
antibodies, the challenged plants were graft inoculated similarly with a T66a severe
isolate.
Decline index is the average per treatment of the cumulative score given by visual
readings on three criteria: stem diameter, growth reduction, and foliage decline
symptoms, where 0 = healthy vigorous plant to 3 = severely diseased. Minimum index
= 0, maximum index =9. A high decline index sometimes was accompanied by plant
death.
2/

24
T66a averaged a decline index of 4.0 (Table 2.5). The lowest
number of dead plants in the experiment (0/5) was obtained
when T26 was the protecting isolate (Table 2.5). At warm
temperatures, Valencia/macrophylla plants inoculated with mild
isolates, including the healthy uninoculated controls gave
decline indexes in the range of 1.7-2.2. In comparison, the
plants pre-inoculated with mild isolates and challenged with
T66a, showed higher decline index values in the range of 2.2-
9. The decline index scores for the control plants
uninoculated with mild isolates but challenged with T66a
averaged 6.8 (Table 2.6). The lowest number of dead plants
occurred when the T26 (0/5) and T55a (0/2) were the protecting
isolates (Table 2.6).
At cool temperatures, the decline index values for
Valencia/sour orange plants inoculated with mild isolates were
in the range of 1.0-3.2. The index of 3.0 for the healthy
uninoculated controls indicated the generally reduced growth
rate of plants at cool temperatures. The decline index values
for plants pre-inoculated with mild isolates and challenged
with T66a ranged from 5.0 to 7.5. The corresponding control
plants uninoculated with mild isolates but challenged with
T66a averaged a decline index of 7.5 (Table 2.7). The lowest
number of dead plants occurred when T26 (0/2) and T55a (1/4)
were the protecting isolates (Table 2.7).
Also at cool temperatures, the Valencia/macrophylla
plants inoculated with T26 and T55a isolates and the

Table 2.6 Effect of citrus tristeza virus (CTV) mild isolates on the development of
the CTV decline syndrome in plants unchallenged and challenged by the T66a severe
isolate: II. Warm temperature, Valencia/macrophylla.
Unchallencred1/
Challencred
1/2/
CTV
No. of
Decline
index/
No. of
Decline
No. of
dead
isolate
plants
plants
index
plants
after
10 m<
challe
Tila
1
2.0
3
9
3/3
T26
4
1.7
5
4.2
0/5
T30
NE-/
NE
NE
NE
NE
T55a
2
2.0
2
2.2
0/2
Healthy 5
or control
plants uninoculated
with mild isolate
2.2
6
6.8
3/6
to
/ One year old plants were graft inoculated with leaf pieces under bark flaps on the
stem from donor plants infected with the indicated mild CTV isolates.
/ After verifying virus infection with mild isolates by DAS-ELISA with polyclonal
antibodies, the challenged plants were graft inoculated similarly with a T66a
severe isolate.
3 / . ...
Decline index is the average per treatment of the cumulative score given by visual
readings on three criteria: stem diameter, growth reduction, and foliage decline
symptoms, where 0 = healthy vigorous plant to 3 = severely diseased. Minimum index
= 0, maximum index =9. A high decline index sometimes was accompanied by plant
death.
4/
NE = treatment not evaluated.

Table 2.7 Effect of citrus tristeza virus (CTV) mild isolates on the development of
the CTV decline syndrome in plants unchallenged and challenged by the T66a severe
isolate: III. Cool temperature, Valencia/sour orange.
Unchallenged1/
Challenqed1/2/
CTV
No. of
Decline
index'
No. of
Decline
No. of dead
isolate
plants
plants
index
plants 10 months
after challenge
Tila
5
3.2
4
7.5
3/4
T26
1
1.0
2
5.5
0/2
T30
2
3.0
3
6.6
2/3
T55a
4
2.7
4
5.0
1/4
Healthy 1
or control
plants uninoculated
with mild isolate
3.0
2
7.5
1/2
/ One year
old plants were graft
inoculated
with leaf
pieces under bark flaps
stem from
i donor
plants infected
with the
indicated
mild CTV isolates.
2 / ,
After verifying virus infection with mild isolates by DAS-ELISA with polyclonal
antibodies, the challenged plants were graft inoculated similarly with a T66a
severe isolate.
to
o>
Decline index is the average per treatment of the cumulative score given by visual
readings on three criteria: stem diameter, growth reduction, and foliage decline
symptoms, where 0 = healthy vigorous plant to 3 = severely diseased. Minimum index
= 0, maximum index =9. A high decline index sometimes was accompanied by plant
death.
2/

27
uninoculated healthy controls showed decline index scores of
5.0, 4.3, and 3.0, respectively. The plants pre-inoculated
with the T26 and T55a isolates and challenged with T66a showed
scores of 4.2 and 3.5, respectively. The uninoculated
controls challenged with T66a showed a decline index of 5.4
(Table 2.8). No dead plants were scored in this experiment
10 months after the challenge inoculations (Table 2.8).
Phloem necrosis at the bud union was not observed in any
treatment at either temperature at five or ten months after
the challenge inoculation (not shown).
Discussion
In this study four different naturally occurring Florida
mild CTV isolates were evaluated for their cross-protecting
ability against the development of the CTV-induced decline in
two susceptible scion/rootstock combinations. DAS-ELISA with
polyclonal antisera was used to determine the total antigen
titer in plants inoculated with mild isolates and those also
challenged with the severe T66a isolate. The MCA-13
monoclonal antibody was evaluated in DAS-indirect ELISA for
quantitation of the T66a severe challenge isolate in mixed
infections.
There were some limitations in the transmissibility of
CTV by leaf piece grafts from the different hosts used to
propagate the CTV isolates. At the beginning of the work most
of the CTV isolates had been propagated in Citrus excelsa
plants, which has been reported as an excellent propagation

Table 2.
the CTV
isolate:
8 Effect of citrus tristeza virus (CTV) mild
decline syndrome in plants unchallenged and
IV. Cool temperature, Valencia/macrophylla.
isolates on the development of
challenged by the T66a severe
Unchallenged1/
Challenged1/2/
CTV
No. of
Decline
index/
No. of
Decline
No. of dead
isolate
plants
plants
index
plants 10 months
after challenge
Tila
NE-/
NE
NE
NE
NE
T26
4
5.0
5
4.2
0/5
T30
NE
NE
NE
NE
NE
T55a
3
4.3
4
3.5
0/4
Healthy 3
or control
3.0
5
5.4
0/5
plants uninoculated
with mild isolate
CO
/ One year old plants were graft inoculated with leaf pieces under bark flaps on the
stem from donor plants infected with the indicated mild CTV isolates.
-/ After verifying virus infection with mild isolates by DAS-ELISA with polyclonal
antibodies, the challenged plants were graft inoculated similarly with a T66a
severe isolate.
3 /
' Decline index is the average per treatment of the cumulative score given by visual
readings on three criteria: stem diameter, growth reduction, and foliage decline
symptom, where 0 = healthy vigorous plant to 3 = severely diseased. Minimum index
= 0, maximum index =9. A high decline index sometimes was accompanied by plant
death.
4/
NE = treatment not evaluated.

29
host for purification purposes of CTV (Lee, et al. 1987b,
1988b) However, it had never been used for leaf graft
transmission experiments. At least three inoculations with
C. excelsa tissue infected with mild isolates were made
unsuccessfully, even though the inoculated tissue survived for
at least four weeks post-inoculation and was left in place
for several months. It was necessary to switch to Madam
Vinous sweet orange and/or Mexican lime plants infected with
the mild isolates as inoculum sources before a minimum of 3
or 4 plants per treatment were positively infected with mild
isolates as determined by DAS-ELISA. Furthermore, some
treatments, mostly Valencia/macrophylla, were not evaluated
because of the lack of replications because not enough plants
were available. This originated the need to design another
separate experiment, described in Chapter 3, to determine the
effectiveness of different citrus species as donor hosts for
graft transmission of the virus.
At warm temperatures, plants inoculated with mild
isolates but unchallenged with T66a had relatively low OD405
values in the range of 0.095-0.145 (Tables 2.1 and 2.2). At
cool temperatures, with few exceptions, were commonly higher
in the range of 0.300-0.400 (Tables 2.3 and 2.4). At warm
temperatures there were some treatments that gave OD405 values
lower than 0.100, which may be interpreted as negative
reactions. However, those values were the averages of the
OD405 readings of every treatment. Thus a single plant with a

30
very low OD405 value could cause the whole treatment average to
be lower than 0.100. It has been suggested that a high titer
in a plant infected with a mild CTV isolate may be a relative
estimate of the protecting ability of mild isolates in cross
protection experiments (Koizumi and Kuhara 1984; Lee et al.
1987a).
Some differences were found in the reaction of the MCA-
13 monoclonal antibody in the different treatments and
temperatures evaluated. At warm temperatures plants pre
inoculated with mild isolates but unchallenged, gave low OD405
values in the range of the uninoculated control plants (Table
2.1 and 2.2). Likewise, at cool temperatures, the OD405 values
obtained with the MCA-13 with mild isolates were slightly
higher than those obtained at warm temperatures (Tables 2.3
and 2.4). This could be interpreted that the MCA-13
monoclonal antibody may react to some extent with mild
isolates when they are above a certain titer in the plants.
However, the OD405 values were always lower than 0.100, which
was considered a negative reaction.
When plants pre-inoculated with mild isolates and further
challenged with the T66a severe isolate were analyzed with the
MCA-13 in DAS-indirect ELISA (Tables 2.1 and 2.2), the OD405
values were generally lower than the unprotected challenged
control plants, even though the differences usually were not
statistically significant (with the exception T55a in Table
2.2) At cool temperatures, a similar phenomenon was observed

31
(Tables 2.3 and 2.4). In this situation, the T26 and T30
isolates generally gave the lowest OD405 values of all
treatments evaluated. This suggests that the mild isolates,
especially T26 and T30, but also T55a to some degree,
prevented or reduced the multiplication of the T66a challenge
isolate. Thus these mild isolates were apparently working as
cross-protecting agents.
These results provide further evidence of the usefulness
of the MCA-13 monoclonal antibody to detect the presence of
severe CTV isolates in mixed infections. The MCA-13 has been
previously used in other studies to evaluate the presence of
severe isolates in field cross-protection experiments (Rocha-
Pea et al. 1990; Yokomi et al. 1990). Determination of the
OD405 readings when the MCA-13 monoclonal antibody is used to
detect the presence of the severe isolate in cross-protection
experiments provides a measurable parameter to estimate the
ability of mild isolates to prevent the establishment of
severe challenge isolates in such experiments.
Some differences were found in the effects of CTV mild
isolates on the inoculated plants and in their ability to
prevent detrimental effects caused by the T66a challenge
isolate at different temperatures. At warm temperatures, the
effect of the CTV mild isolates on performance of both
Valencia/sour orange and Valencia/macrophylla were negligible.
The decline indexes for the healthy uninoculated control
plants were nearly egual to or greater than those inoculated

32
only with mild isolates (Table 2.5 and 2.6). In regard to
Valencia/sour orange plants pre-inoculated with mild isolates
and challenged with the T66a severe isolate, the lowest
decline index (2.6) and lowest number of dead plants (0/5)
was obtained with the T26 isolate. Whereas, the highest
decline index (6.3) and highest number of dead plants (3/8)
was obtained with the Tila isolate. A decline index of 4.0
and 1/6 dead plants were scored with the uninoculated and
challenged control plants (Table 2.5). Also at warm
temperatures, Valencia/macrophylla plants pre-inoculated with
mild isolates and challenged with the T66a severe isolate, the
T26 and T55a isolates obtained the lowest decline index (4.2
and 2.2) and lowest number of dead plants (0/5 and 0/2),
respectively.
At cooler temperatures, there were some differences in
the effect of the CTV mild isolates on growth of both
Valencia/sour orange and Valencia/macrophylla. The decline
indexes for the healthy uninoculated control plants were
variable and ranged from values below to above those obtained
with plants inoculated only with mild isolates (Table 2.7 and
2.8). In general, there was a remarkable growth reduction
effect at cool temperatures even on healthy uninoculated
control plants. In this regard Valencia/sour orange plants
pre-inoculated with mild isolates and challenged with the T66a
severe isolate, the T26 and T55a isolates obtained the lowest
decline index (5.5 and 5.0) and lowest number of dead plants

33
(0/2 and 1/4), respectively. Whereas, the highest decline
index (7.5) and highest number of dead plants (3/4) were
obtained with the Tila isolate. A decline index of 7.5 and
1/2 dead plants were scored with the uninoculated and
challenged control plants (Table 2.7). At cool temperatures,
Valencia/macrophylla plants pre-inoculated with mild isolates
and challenged with the T66a severe isolate, had decline index
scores similar to the plants inoculated only with mild
isolates. No dead plants were obtained in this portion of the
experiment (Table 2.8).
Of all treatments evaluated at both temperatures, the
T26 isolate, and also T55a to some degree, obtained the lowest
decline index scores and lowest number of dead plants as
compared with the uninoculated challenged control plants.
This provided further evidence of their cross-protecting
effect, especially the T26 isolate against the development of
the CTV-ID syndrome.
Of special interest was the high decline index scores
and number of dead plants in plants pre-inoculated with the
Tila mild isolate and further challenged with the T66a
isolate. It seemed that the combination of Tila and the T66a
isolates produced a more severe reaction on the challenged
plants than that caused by the T66a isolate alone in the
unprotected control plants. The lack of cross-protecting
ability of the Tila isolate has been previously reported
(Yokomi et al. 1987).

34
The relatively low decline index scores and low
occurrence of dead plants at both temperatures in the
unprotected control plants challenged with the T66a isolate,
indicates that under greenhouse conditions, the use of one
single challenge isolate might not be sufficient to obtain an
appropriate rate of decline in a short time basis. It is well
documented that CTV occurs naturally as mixtures of isolates
or strains with diverse biological properties (Garnsey et al.
1987, McClean, 1974). The T66a severe isolate was originally
isolated from an infected field source, and subsequently aphid
transmitted to avoid contamination with other viruses (Garnsey
et al. 1987; Yokomi and Garnsey, 1987) It is a possibility
that part of the original decline components from the field
source could have been lost in the subsequent aphid
transmissions. To overcome this possibility, it may be
advisable in the future to use a mixture of several severe
isolates as a challenge to enhance the possibility of
obtaining an appropriate occurrence of decline under
greenhouse conditions. Another alternative could be the use
of higher populations of aphids (50 or 100) to obtain a more
complete complex of CTV severe isolates from field samples.
Cross-protection using mild virus isolates as a strategy
to reduce losses due to CTV has been used in Brazil (Costa and
Muller, 1980; Muller, 1980), South Africa (DeLange et al.
1980; Garnsey and Lee, 1988), Japan (Ieki, 1989; Koizumi,
1986), India (Balaraman and Ramakrishnan, 1980) and Australia

35
(Cox et al. 1976; Fraser et al. 1968), against stem-pitting
isolates either in orange, grapefruit, and/or acid lime.
Several approaches have been reported for the evaluation of
mild isolates under greenhouse conditions. However, these
approaches have been addressed mostly to the evaluation of the
cross-protecting effect of mild isolates against stem pitting
and have included only the host reaction of Mexican lime,
sweet orange, or grapefruit seedlings (Roistacher et al. 1987,
1988; Van Vuuren and Noll, 1987). Another approach where the
challenge inoculations are made by using insect vectors to
screen mild isolates (Yokomi et al. 1987) has not been
extensively used.
The results of this work provide further evidence that
a) the cross-protection against the CTV-induced decline on
sweet/sour orange combinations may be possible; b) the
preliminary evaluation of mild isolates under greenhouse
conditions can be made in a relatively short time basis, and
c) the severe challenge isolate can be detected by using the
MCA-13 strain specific monoclonal antibodies. The recent
report of Miyakawa (1987) about the feasibility of cross
protection on sweet/sour orange combinations, supports these
conclusions.
Considering the relatively high virus titer found at cool
temperatures, it is advisable to propagate the donor plants
at temperatures in the range of 21-33C to better guarantee a
high percentage of CTV transmission to the receptor plants.

36
Likewise, a mixture of several severe isolates should be used
as the challenge virus source to give a better evaluation of
decline symptoms under greenhouse conditions.
The methodology described herein offers the following
advantages: i) Depending upon space availability, large
numbers of mild isolates can be evaluated uniformly in a time
period of 18 to 24 months: from six to twelve months to get
the one-year-old plants infected with the mild isolates and
verification of infection by serology, two months for
challenge and ten months for final evaluation; ii) The
availability of the MCA-13 monoclonal antibody provides a
useful tool to detect the presence of a severe isolate in the
challenged plants, and at the same time allows an estimate of
the relative ability of mild isolates to prevent the
establishment of the severe isolate in the challenged plants;
iii) Mild isolates can be evaluated in grenhouses without the
risks that represent the threat of recurrent freezes
especially in Florida in recent years, the lack of an
appropriate natural challenge pressure (vector or severe
isolate), and the effect of some other devastating diseases
(i.e. greening or blight) that can hamper the reliable
evaluation of cross-protection experiments under field
conditions. Some limitations in the methodology can be also
visualized. The use of leaf piece grafts is not always highly
efficient to transmit CTV from the donor propagation hosts to
the receptor test plants (see Chapter 3) This leaves the


37
possibility that a lack of transmissibility by leaf piece
grafts of the CTV challenge isolate, can be interpreted
erroneously as protecting effect by mild isolates. On the
other hand, the inoculum tissue with the severe isolate is
left in place to enhance the probability of graft-transmission
in the challenged plants. This would supply a permanent
source of the severe isolate against the mild isolates which
may provide a stronger challenge pressure than happens under
natural conditions. If this occurs, mild isolates with
potential protecting ability under natural challenge
conditions could be underestimated or overlooked.

CHAPTER 3
EFFECTIVENESS OF CITRUS SPECIES AS DONOR HOSTS FOR
GRAFT TRANSMISSION OF CITRUS TRISTEZA VIRUS
Introduction
Citrus tristeza virus (CTV) has long been known to be
transmitted by budding and by different grafting procedures
(Bennett and Costa, 1949; Bar-Joseph et al. 1979a; Bar-Joseph
and Lee, 1990). In 1951, Wallace experimentally transmitted
CTV by placing small portions of donor leaf or bark tissue
under a flap of bark on receptor plants. By this method, CTV
was transmitted from many field sources of sweet orange to
Mexican lime receptor plants, and from Mexican lime to healthy
sweet orange plants (Wallace, 1951). Schwartz (1968)
transmitted CTV by connecting the distal portion of the leaf
of infected plants to a matching proximal part on a leaf of
a receptor plant. By this method, CTV was transmitted to 9
of 20 Mexican lime plants, but transmission was obtained only
when callus formation occurred between grafted tissues; also,
older and dark green leaves were a better source than younger
leaves for both callus formation and virus transmission.
Cohen (1972) described a method for CTV transmission in
citrus by grafting triangular leaf pieces into triangular
holes cut in the leaves of receptor plants. He transmitted
38

39
CTV from Meyer lemon to 25 of 27 Mexican lime and sour orange
seedlings when leaf pieces contained midribs. However, the
efficiency of transmission decreased more than 50% when grafts
did not include the leaf midrib. A modification of Cohen's
procedure, called "leaf-disc grafting" (Blue et al. 1976)
involved the use of circular leaf pieces 6 mm in diameter cut
from the midrib area of a donor plant leaf and placed into a
corresponding hole in the receptor plant leaf. The midrib of
the donor tissue is aligned with that of the receptor leaf,
and grafts are held in place with transparent tape. This
method was as successful as bud inoculation for transmitting
many CTV isolates to Mexican lime plants from different citrus
species, and was more efficient for transmitting mild CTV
isolates. The leaf-disc method was used for routine indexing
in the citrus budwood certification program in California
(Calavan et al. 1978).
Another method involving the use of leaf piece grafts
was reported by Garnsey and Whidden (1970). Rectangular leaf
pieces from infected plants were inserted under corresponding
s .
rectangular bark flaps cut in the stem of receptor hosts.
This procedure has been used widely for many years with CTV
and other citrus viruses, and it has been used in the
characterization of the biological properties of diverse
worldwide collection of CTV isolates (Garnsey et al. 1987).
Leaf piece grafts are especially advantageous when large

40
numbers of plants are to be inoculated with limited sources
of inoculum (Garnsey and Whidden, 1970).
During several experiments with CTV in Florida (Chapter
2; Rocha-Pea et al. 1990) large numbers of plants were
inoculated by leaf piece grafts with several CTV isolates that
were propagated in different citrus hosts. There were notable
differences in the efficiency of transmission of some CTV
isolates from different donor hosts, and in some cases no
transmission was achieved even after repeated inoculations.
The objectives of this research were to evaluate the effect
of different citrus hosts on the efficiency of graft
transmission of CTV, and to determine the relative
distribution of the virus in different host tissues.
Materials and Methods
Virus isolates and donor hosts. Three isolates of CTV,
T26, T30, and T66a, were used throughout the study. They have
been described previously (Garnsey et al. 1987; Lee, 1984;
Yokomi and Garnsey, 1987). Virus isolates were propagated in
Citrus excelsa Wester, Mexican lime {C. aurantifolia
(Christm.) Swingle} and Madam Vinous sweet orange {C. sinensis
(L.) Osb.} plants, herein referred to as donor hosts,
maintained in a greenhouse with mean minimum and maximum
temperatures of 21 and 33C, respectively. Inoculum tissue
from donor hosts was evaluated by serological indexing by the
double antibody sandwich enzyme-linked immunosorbent assay
(DAS-ELISA) (see below) to verify the presence of CTV before

41
being used as inoculum. All donor host plants were known to
be CTV infected for at least one year before the study was
started.
Grafting procedures and receptor hosts. Rectangular leaf
and bark pieces of about 3 X 15 mm were cut from donor hosts
with a sharp knife and inserted under corresponding bark flaps
cut on the stem of one-year-old Madam Vinous sweet orange,
Mexican lime, and grapefruit (C. paradisi Macf.) plants,
herein referred to as receptor hosts. A portion of the
grafted tissue (2-3 mm) was left exposed at the top of bark
flaps to monitor tissue survival at 21 days post-inoculation.
A minimum of five plants of each receptor host were each
inoculated with 4 pieces of either leaf or bark tissue for
every donor host/virus isolate combination tested.
Serological indexing by the double antibody sandwich enzyme-
linked immunosorbent assay (DAS-ELISA) (see below) was carried
out on receptor hosts at three and five months post
inoculation.
Inoculated receptor plants were grown in a commercial
potting mixture (Pro-mix BX) in three liter plastic
containers, and fertilized with a mixture of NPK (20-10-20)
every other week, and given disease and pest management as
described in Chapter 2.
Virus distribution and antigen concentration in host
tissues. Individual Madam Vinous sweet orange and C. excelsa
plants infected with CTV isolates T26 or T66a were used to

42
study the relative distribution and antigen concentration of
the virus in different tissues of the host plant. Bark,
petioles, midribs, and leaf blades of four individual branches
of each test plant, were assayed individually by DAS-ELISA.
At least four replications were assayed for every host/virus
isolate combination tested.
Purification of CTV. Citrus tristeza virus was purified
from tender new tissue of C. excelsa greenhouse grown plants
infected with the T26 isolate, by the Driselase method
(Garnsey et al. 1981b; Lee et al. 1988b). The final virus
preparations were adjusted with 0.05 M Tris buffer to optical
density values (OD260) of 0.4 and stored in one ml aliquots at
18C.
Serological tests. The double antibody sandwich enzyme-
linked immunosorbent assay (DAS-ELISA) (Bar-Joseph et al.
1979b, 1980) was conducted with polyclonal antiserum no. 1053
prepared against whole, unfixed CTV isolate T26 (R.F. Lee,
unpublished). Polystyrene Immulon II microtiter plates
(Dynatech Laboratories) were used. Unless otherwise stated,
200 microliters were used per well of the microtiter plates
and, three washings with phosphate buffered saline (PBS)-Tween
{PBS = 8 mM Na2HP04, 14 mM KH2P04, 15 mM NaCl, pH 7.4, (+0.1
% Tween 20)> were performed between steps. Host tissue (bark,
petioles, midribs, etc.) was chopped finely with a razor blade
and ground in a Tekmar Tissumizer in extraction buffer (PBS-
Tween + 2% polyvinyl pyrrolidone (PVP-40 Sigma) at a 1:20

43
(w/v) dilution. Microtiter plates were coated with 2.0 jug/ml
of purified CTV specific IgG in carbonate buffer (0.015 M
NaHC03, 0.03 M NaC03, pH 9.6) and incubated for 6 hr at 37C.
Antigen samples were added to the wells and incubated for 18
hr at 5C. CTV specific IgG conjugated to alkaline
phosphatase was used at a dilution of 1:1,000 in conjugate
buffer (PBS-Tween + 2% PVP + 0.2% bovine serum albumin) and
incubated for 4 hr at 37C (Bar-Joseph et al. 1979b, 1980).
The reaction with one mg/ml of p-nitrophenyl phosphate (Sigma)
in 10% triethanolamine, pH 9.8, was measured at 120 min at 405
nm (OD405) with a Bio-Tek EL-307 ELISA plate spectrophotometer.
Samples were considered positive when OD405 values were higher
than 0.100 or three times the mean of healthy controls,
whichever was greater. There were two replications per sample
in each microtiter plate. To estimate the relative CTV
concentration in test samples, a standard curve prepared by
diluting purified CTV T26 to OD260 values of 0.04, 0.02, 0.01,
0.005, 0.0025, 0.00125, and 0.0006 in a PBS-Tween + PVP
buffered extract of bark of healthy Citrus excelsa it was
included as a positive control in every test. Negative
controls included PBS-Tween + 2% PVP, conjugate buffer, and
extract from healthy C. excelsa. Madam Vinous sweet orange,
Mexican lime and grapefruit plants.
Results
Graft transmission of citrus tristeza virus isolates.
At 21 days post-inoculation the survival rate of grafted

44
tissue in the whole experiment was 83 and 66 percent for leaf
and bark pieces, respectively. Overall at least one of the
four grafts survived in 92 and 90 percent of the receptor
plants inoculated with leaf or bark pieces, respectively. In
calculating the percent of virus transmission for each
donor/receptor host/virus isolate combination, only those
plants with at least one (of four) surviving inoculum piece
were taken into account. Thus, overall there was a greater
efficiency of transmission with leaf pieces (89.2%) than with
bark pieces (75.6%) for the whole experiment (Table 3.1).
There were three plants of 270 in the entire experiment, one
Mexican lime and two grapefruit that became infected even
though no successful graft was scored 21 days post
inoculation.
The overall rate of transmission of CTV by graft
inoculation for each donor/receptor host combination is shown
in Table 3.2. With C. excelsa as donor host there was 72.4%,
86.9%, and 60.7% transmission to Madam Vinous, Mexican lime
and grapefruit, respectively. With Mexican lime as donor
host, there was 93.1%, 76.9% and 89.3% transmission to Madam
Vinous, Mexican lime and grapefruit, respectively. With Madam
Vinous as donor host there was 86.7%, 100%, and 84.6%
transmission to Madam Vinous, Mexican lime and grapefruit
respectively. The overall average of transmission was 72.5%
from C. excelsa. 85.2% from Mexican lime, and 90.6% from Madam
Vinous (Table 3.2). Statistical analysis showed significant

45
Table 3.1 Transmission of citrus tristeza virus by graft
inoculation between selected citrus hosts: I. Efficiency of leaf
and bark pieces as inoculum.
Inoculum
tissue
Inoculum
survival%-/
% plants with
at least one % transmission-7
successful graft
leaf
8 3. O^a^7
92.0
89.2 a
bark
66.0 b
90.0
75.6 b
Measured at 21 days post-inoculation.
Percent transmission to plants with at least one inoculum
piece (of four) alive, measured serologically by DAS-ELISA at
3 and 5 months post-inoculation. Number indicates overall
transmission for all donor/receptor/virus isolate combinations.
A total of 270 plants (135 each) were inoculated with four pieces
of either leaf or bark tissue. Number indicates overall survival
for all donor/receptor/virus isolate combinations.
Numbers in the same column followed by different letters are
statistically different by Duncan's test (P < 0.05).

Table 3.2 Transmission of citrus tristeza virus by graft inoculation between selected citrus
hosts: II. Overall rate of transmission.
Receptor host
Donor host
Madam Vinous
Mexican
lime
Grapefruit
Average
Citrus excelsa
72.4-/-/b^/
86.9
ab
60.7 b
72.5-/b
Mexican lime
93.1 a
76.9
b
89.3 a
85.2 ab
Madam Vinous
86.7 ab
100.0
a
84.6 a
90.6 a
4.
CTi
/ Percent transmission to plants with at least one inoculum piece (of four) alive 21 days
post-inoculation. Number indicates overall of transmission for all virus isolates
combinations.
2 / ,
Each value represents a minimum of 27 plants.
3 / ...
' Numbers in the same column following by different letters are statistically different
by Duncan's tests (P < 0.05).
4 / , ,
Number indicates the overall transmission for all receptor/virus isolate combinations.

47
differences (P < 0.05%) for all donor-receptor host
combinations. Likewise, according to Duncan's multiple range
comparison test, there were statistical differences between
some of the hosts tested (Table 3.2).
The rate of transmission for the three different CTV
isolates tested with each donor host is shown in Table 3.3.
The T26 isolate was transmitted at a rate of 69.0% to 92.8%.
The transmission rates of T30 and T66a isolates ranged from
62.5% to 100% and from 71.4% to 96.7%, respectively, from the
hosts tested. While there were statistical differences in
the rates of transmission for some of the virus isolate/donor
host combination, the overall average of transmission showed
no significant differences among them (Table 3.3).
The overall statistical analysis for percent transmission
of the interactions among the different donor/receptor/virus
isolate/inoculum pieces combinations, indicated no significant
differences for receptor and virus isolates alone, and for the
combinations of donor/inoculum pieces, receptor/inoculum
pieces, and for donor/receptor/virus isolate. However,
significant differences (P < 0.05) were found for donor and
inoculum pieces alone, and for the interactions between
donor/receptor, donor/virus isolate, receptor/virus isolate,
and virus isolate/inoculum pieces.
Virus distribution and antigen concentration in host
tissues. The relative antigen titer of CTV as measured by
DAS-ELISA in each host tissue/virus isolate combination is

Table 3.3 Transmission of citrus tristeza virus by graft inoculation between selected citrus
hosts: III. Effect of virus isolates.
Donor host
Virus
isolate
Citrus
excelsa
Mexican
lime
Madam Vinous
sweet orange
Averaae
T26
69.0-/-/a-/
89.3
a
92.8 a
83.5-/a
T30
76.7 a
65.2
b
100.0 a
80.7 a
T66a
71.4 a
96.7
a
77.3 b
83.5 a
. 03
/ Percent transmission to plants with at least one inoculum piece alive 21 days post
inoculation. Number indicates overall of transmission for all donor/receptor host
combinations.
2 / .
' Each value represents a minimum of 27 plants.
3 / ...
Numbers in the same column following by different letters are statistically different
by Duncan's tests (P < 0.05).
4 / ,
Number indicates overall transmission for all donor/receptor host combinations.

49
illustrated in Table 3.4. The average optical density values
at 405 nm (OD405) for bark tissue were 0.221 and 0.349 for the
T26 isolate and 0.266 and 0.336 for the T66a isolate in Madam
Vinous and C. excelsa. respectively. The OD405 values found in
the other tissues assayed in both hosts for T26 and T66a
isolates were in the range of 0.137 and 0.188 and 0.099 and
0.238 for petioles, 0.173 and 0.241 and 0.049 and 0.133 for
midribs, and 0.044 and 0.065 and 0.030 for leaf blades,
respectively. There were significant statistical differences
between C. excelsa and Madam Vinous for bark tissue with the
T26 isolate, and for both petioles and midribs with the T66a
isolate. The overall analysis showed that the highest OD405
values in both hosts for both T2 6 and T66a isolates, were
found in bark, followed by petioles and midribs. Leaf blades
showed the lowest OD405 values of all tissues assayed in both
hosts and isolates tested. Some differences in the OD405
values were found between different parts of the same plant,
and from one plant to another, in some virus isolate/host
combinations; however, the statistical analysis did not show
significative differences among them (data not shown). From
the standard curve prepared with purified T26 (Fig. 3.1), it
was estimated that an OD405 value of 0.4 65 was approximately
equivalent to 20 /g/ml of CTV, assuming an extinction
coefficient of 2.0 (Gonsalves et al. 1978). Therefore, the
CTV antigen concentration in the test samples (10 mg of

Table 3.4 Relative antigen titer of citrus tristeza virus in different tissues of Citrus
excelsa and Madam Vinous sweet orange host plants, as measured by enzyme-linked immunosorbent
assay.
Host tissue
Virus
isolate
Donor host
Bark
Petioles
Midribs
Leaf
blade
T26
Citrus excelsa
0.349-^/a-/
0.137 a
0.086 a
0.044
a
Madam Vinous
0.221 b
0.188 a
0.120 a
0.065
a
T66a
Citrus excelsa
0.336 a
0.238 a
0.133 a
0.030
a
Madam Vinous
0.266 a
0.099 b
0.040 b
0.029
a
/ Mean of optical density (OD4Q5) per 10 mg plant tissue after 120 min of substrate reaction.
There were four replicates per plant and four plants per host/isolate combination.
Control reaction with the same tissue from healthy plants averaged OD4Q5 = 0.001-0.025.
This has not been subtracted from the values above.
! Numbers in the same column per host/isolate combination followed by different letters
are statistically different by Duncan's test (P < 0.05).

51
Purified virus (Optical Density 260 nm)
Figure 3.1 Plot of purified citrus tristeza virus (CTV)
against optical density. Bark of healthy Citrus excelsa (0.25
g) tissue was ground in 5.0 ml of phosphate buffered saline,
pH 7.6, + 0.05% Tween + 2% polyvinyl pyrrolidone and mixed
with purified CTV T2 6 isolate to give the desired optical
density at 260 nm (OD260) DAS-ELISA was performed as described
in materials and methods. An extinction coefficient of 2.0
was assumed (Gonsalves et al. 1978) to estimate the relative
virus concentration.

52
tissue/200 /xl) ranged from an average of 0.2-0.5 ng in leaf
blades to 1.9-2.9 jug in bark tissue.
Discussion
In this study three CTV isolates were graft-transmitted
by using leaf or bark tissue from three citrus donor hosts to
three receptor hosts. The simple establishment and survival
of grafted tissue in the receptor host was not sufficient to
transmit CTV from certain donor hosts. There were a number
of instances, 22 of 80, and 12 of 81, respectively, when C.
excelsa and Mexican lime were used as donor hosts, where no
transmission was achieved even on those receptor plants where
at least one grafted tissue piece was still alive 21 days
post-inoculation. Similar results were obtained, but to a
lesser degree (7 of 75) when Madam Vinous sweet orange was the
donor host. Furthermore, some of the receptor plants where
no transmission was scored had all four grafted pieces still
alive even five months post-inoculation.
The overall analysis of the results showed significant
differences in the efficiency of the three donor hosts tested
to transmit CTV (Table 3.2). Likewise, differences were found
in the rate of transmission for each donor/receptor host
combination. For example, C. excelsa showed rates of
transmission of 72.4% and 60.7% to Madam Vinous and
grapefruit, respectively; whereas, a rate of transmission of
86.7% was obtained to Mexican lime plants. In regard to
Mexican lime as donor host, there was a rate of 89.3% and

53
93.1.7% transmission to grapefruit and Madam Vinous,
respectively and a 76.9% rate to Mexican lime. Transmission
from Madam Vinous sweet orange was between 84.6 and 100% in
all receptors tested. It was surprising that transmission
rates between the same species were only 86.7% for Madam
Vinous, and 76.9% for Mexican lime (Table 3.2).
Madam Vinous sweet orange was the most efficient donor
host with the three receptor hosts tested (90.6%), followed
by Mexican lime (85.2%). C. excelsa was a poor donor host
(72.5%), being relatively efficient only when inoculated to
Mexican lime (Table 3.2).
Previous studies on the transmission of CTV by grafting
procedures have shown that a period of at least ten days
contact between grafted tissues is needed to obtain
transmission of the virus to the receptor host (Tolba et al.
1976; Yamaguchi and
Patpong,
1980). In
this
study,
the
survival of
grafted
tissue
was scored
21
days after
inoculation,
but the
inoculated tissue
was left in
the
receptor plants for up to five months. This should have been
ample time for contact between the inoculum and receptor
cambium to establish a tissue union with a subsequent
transmission of CTV. Furthermore, when leaf pieces were used
as inoculum, a small portion of the midrib was included in
every piece to increase the success of the grafting. The
overall rates of successful grafts were about 83% and a 66.7%
for leaf and bark pieces, respectively (Table 3.1).

54
A,
The reason why a low percentage of graft transmission of
the virus was found from some donor hosts, and the absence of
an expected 100% when the donor-receptor combination was of
the same species, is unknown. A possible explanation could
be differences in the virus distribution and/or concentration
in the donor tissues used as inoculum. Bark tissue contained
the highest antigen titer with OD405 values in the range of
0.221 and 0.349 in both C. excelsa and Madam Vinous with both
CTV isolates tested (Table 3.4). These values were, in some
instances, more than double those found in petioles and
midribs, and at least triple those found in the leaf blade.
Even though the statistical analysis did not show significant
differences in antigen titer in either different parts of the
same plant or from one plant to another, there were some
instances where OD405 values were as low as the healthy
controls. This indicates a possible absence of the virus in
those tissues and raises the possibility that occasionally
the tissue used for graft transmission may be virus-free, with
a subsequent failure in the transmission. Other possibilities
could be an occasional absence of phloem connections between
the donor and receptor tissues with a subsequent absence of
movement of the virus across the junction or the requirement
of a minimum of virus particles present in the tissue used as
inoculum in order to accomplish the transmission.
Citrus tristeza virus is phloem-limited (Bar-Joseph et
al. 1979a; Lister and Bar-Joseph, 1981), and is normally found

55
at higher concentrations in young phloem-rich tissues
(Garnsey, et al. 1979; Bar-Joseph, et al. 1979a); however,
the serological titer frequently decreases as the tissues
reach maturity or when the plants are exposed to warm
environments (Garnsey et al. 1981a; Lee et al. 1988c). The
OD405 values obtained in this research were low if compared
with those found when DAS-ELISA is used routinely for CTV
diagnosis (Bar-Joseph et al. 1979b; Garnsey et al. 1980b);
however, this part of the work was addressed to determine the
virus titer in the tissues suitable for graft transmission,
and young tender tissue sometimes is not a good source of
inoculum for leaf piece grafts (personal observations).
The overall analysis of the results obtained indicates
that the efficiency of the graft transmission of CTV is
conditioned primarily by the donor/receptor host combination,
and secondly by the virus isolate involved, but apparently not
by the interaction of the three. For example, C. excelsa
showed an overall rate of transmission in the range of 72.5%
with all receptor hosts tested (Table 3.2), and a similar low
pattern between 69% and 76.7% (= 72.8%) was obtained for the
three isolates tested (Table 3.3). Likewise, when Madam
Vinous was used as the donor host, there was an overall rate
of transmission of 90.6% (Table 3.2). A rate of 77.3-100% (=
88.6%) occurred from this host with the three isolates tested
(Table 3) A comparable event was also scored when Mexican
lime was the donor host (Table 3.2 and 3.3). The statistical

56
significance found for the interactions donor/receptor and
donor/virus isolate, and no significance for the interaction
of donor/receptor/virus isolate supports this conclusion.
The use of leaf and/or bark pieces for graft transmission
of CTV may be advantageous when large numbers of plants are
to be inoculated with limited sources of inoculum (Garnsey and
Whidden, 1970). However, in the light of the results of this
research, in order to achieve a high level of transmission,
the efficiency of the donor host and the donor/receptor host
combination should be considered.

CHAPTER 4
DEVELOPMENT OF A DOT-IMMUNOBINDING ASSAY FOR
CITRUS TRISTEZA VIRUS
Introduction
Several serological methods have been developed and used
to diagnose the presence of citrus tristeza virus (CTV) in
infected tissue. These methods include: SDS-immunodiffusion
procedures (Garnsey et al. 1979; Bar-Joseph et al. 1980), in
situ immunofluorescence (Brlansky et al. 1984; Tsuchizaki et
al. 1978), serologically specific electron microscopy
(Brlansky et al. 1984; Garnsey et al. 1980), gold
immunolabeling microscopy (Davis and Brlansky, 1990), and
several enzyme-linked immunosorbent (ELISA) procedures (Bar-
Joseph and Malkinson, 1980; Bar-Joseph et al. 1979b) including
the use of the biotin-avidin system (Irey et al. 1988) and the
enzyme-amplified ELISA (Ben-Ze'ev et al. 1988) to increase the
sensitivity of detection. Each of these methods has different
advantages and sensitivity levels, and therefore, has been
used for different purposes and applications (Rocha-Pea and
Lee, 1990).
Polyclonal antisera specific for CTV have been developed
in different animal species, such as rabbits (Gonsalves et al.
1978; Tsuchizaki et al. 1978; R.F. Lee, unpublished;), and
57

58
chickens (Bar-Joseph and Malkinson, 1980; Marco and Gumpf,
1990). Likewise, several monoclonal antibodies (MCAs) have
been developed in mice (Gumpf et al. 1987; Permar et al. 1990;
Vela et al. 1986, 1988). The 3DF1 MCA has been reported to
react with a broad spectrum of CTV isolates of different
geographical origins and is available commercially (Vela et
al. 1986, 1988). The MCA-13 MCA (hereafter MCA-13) has been
reported to react specifically with CTV isolates that have
severe biological activities, especially isolates causing
decline on plants grafted on sour orange rootstock (Garnsey
and Permar, 1990; Permar et al. 1990). It has been used in
diverse studies for strain discrimination purposes (Irey et
al. 1988; Garnsey and Permar 1990; Rocha-Pea et al. 1990;
Yokomi et al. 1990).
The ELISA test, particularly the double antibody sandwich
system (DAS-ELISA) (Bar-Joseph et al. 1979b; 1980) has been
the most widely used of all serological methods developed for
CTV detection (Bar-Joseph et al. 1989; Garnsey, et al. 1981a;
Rocha-Pea and Lee, 1991) In DAS-ELISA the virus in the test
sample is trapped and immobilized selectively by specific
antibodies adsorbed on polystyrene microtiter plates. Enzyme-
conjugated antibodies are then reacted with the trapped virus
and detected colorimetrically after adding a suitable
substrate (Clark and Adams, 1977; Garnsey and Cambra, 1990).
DAS-ELISA is relatively easy to perform and is highly
sensitive, but it does require some special equipment, it is

59
laborious and time-consuming for large scale indexing, and
both antibodies and large volumes of buffer are used in each
test.
A simple and rapid serological method, known as dot-
immunobinding assay (DIBA) (Hawkes et al. 1982), has been
developed and applied for the detection of several plant
viruses (Hibi and Saito, 1985; Powell, 1987). The principles
of DIBA are similar to those of ELISA, differing mostly in
that the antigen and antibodies are bound to nitrocellulose
membranes instead of polystyrene microtiter plates, and that
the product of the enzyme reaction at the end of the test is
insoluble. The use of nitrocellulose membranes for the
serological detection of plant viruses has become popular
because of the simplicity of equipment required and lower cost
of materials needed. In a preliminary study the DIBA test was
as sensitive as DAS-ELISA to detect CTV in both field trees
and greenhouse grown plants (Rocha-Pea et al. 1990); however,
some differences were found in the reactivity of the
antibodies used, and some nonspecific reactions occurred in
the test. The objective of this research was to adapt DIBA for
diagnosis of CTV using different polyclonal and monoclonal
antibodies and to compare the sensitivity of DIBA with DAS-
ELISA and DAS-indirect ELISA.
Materials and Methods
Antisera used. Polyclonal antisera numbers 1051, 1052,
and 1053 previously prepared in rabbits against undegraded,

60
and unfixed virus particles of T30, T3 6, and T2 6 CTV isolates,
respectively (R.F. Lee, unpublished) were used.
Immunoglobulins (IgG) were purified from whole sera by
the Protein A-Sepharose affinity chromatography method (Miller
and Stone, 1978) and adjusted to a final concentration of 1.0
mg/ml (OD280= 1.40) in PBS buffer with 0.02% sodium azide and
stored at 4C (Clark et al. 1986) The 3DF1 MCA was a gift
from Drs. P. Moreno and M. Cambra, Valencia, Spain, and its
preparation was described previously (Vela et al. 1986, 1988).
The MCA-13 that reacts specifically with severe CTV strains
(Permar et al. 1990) was a gift from Drs. T.A. Permar and S.M.
Garnsey. Goat anti-mouse and goat anti-rabbit IgG conjugated
with alkaline phosphatase were purchased from either
Boehringer or Promega.
The purified polyclonal IgG and 3DF1 MCA were tested at
concentrations of 1.0, 0.1, 0.2 or 0.01 /xg/ml. The MCA-13
was used as ascites fluid at a dilution of 1:5,000 (v/v) .
Goat anti-species IgG were used at concentrations recommended
by the manufacturer.
Sample preparation.- Bark tissue was peeled from fully-
expanded new flushes of CTV infected and healthy citrus
plants. The tissue was finely chopped and homogenized with
a Tekmar Tissumizer in the presence of Tris buffered saline
(TBS)-Tween (TBS = 0.02 M Tris, 0.5 M NaCl, pH 7.5), plus 0.5
% Tween 20 (TBS-Tween) at 1:10 (w/v) dilution.

61
Dot-immunobindinq assay (PIBA).- The general protocol
used for DIBA was as follows: nitrocellulose membranes (Micro
Separations, Inc.)/ 0.45 m pore, were cut at a size of 11 X
7.5 cm and wet in TBS for at least 30 min, blotted on
chromatography paper (Whatman No. 1), and allowed to dry for
5 min before use. Aliquots of 2 /xl of test samples were
applied to nitrocellulose membranes by using as a guide a
template constructed from the rack of a micropipet tip holder
box, allowed to dry for at least 10-15 min or stored at room
temperature for several days before use. All subsequent
incubation steps were performed at room temperature in 25 ml
of each solution using polypropylene covers of micropipet tip
holder boxes as trays. During the incubation or washing the
membranes were agitated gently in a shaker at 50 oscillations
per min or agitated by hand. Nitrocellulose membranes, with
the test samples, were soaked for 30 min in blocking solutions
of either 10% horse serum (v//v), 3% bovine serum albumin
(BSA) (w/v), 3% gelatin (w/v), 5% Triton X-100 (v/v), or 0.5%
non-fat dry milk (w/v), all in TBS, and washed twice with 25-
50 ml TBS-Tween and once with TBS, 2 min each. Then the
membranes were incubated with the CTV specific antibodies for
either 30 min, one, two, or 18 hr, depending on the IgG used,
and washed again as after blocking. The membranes were then
incubated for 30 min with the corresponding goat anti-species
IgG conjugated with alkaline phosphatase. After washing as
before, the membranes were incubated in the substrate

62
solution. The substrate solution was prepared as follows: 10
mg of nitro blue tetrazolium (NBT) (Sigma) were dissolved in
30 ml of TBS substrate buffer (0.1 M Tris, 0.1 M NaCl, 0.005
M MgCl2, pH 9.5); then 5 mg of 5-bromo-4-chloro-3-indoyl
phosphate (BCIP) (Sigma) were added and dissolved in the
solution. The color reaction was stopped by transferring the
membranes to distilled water.
DAS-ELISA.- The double antibody sandwich enzyme-linked
immunosorbent assay (DAS-ELISA) (Bar-Joseph et al. 1979b,
1980) was conducted using polyclonal IgG No. 1053 and
polystyrene Immulon II microtiter plates (Dynatech
Laboratories). Unless otherwise stated, 200 microliters were
used per well and three washings with PBS-Tween (phosphate
buffered saline = 8 mM Na2HP04, 14 mM KH2P04, 15 mM NaCl, pH
7.4, + 0.05% Tween 20) were performed between steps.
Microtiter plates were coated using 3.0 ng/ml of purified CTV
specific IgG in carbonate buffer (0.015 M NaHC03, 0.03 M NaC03,
pH 9.6) and incubated for 6 hr at 37C. Antigen samples were
added to the wells and incubated for 18 hr at 5C. CTV
specific IgG conjugated with alkaline phosphatase at a
dilution of 1:1,000 in conjugate buffer {PBS-Tween + 2%
polyvinyl pyrrolidone (=PVP 40,000 MW) (w/v) + 0.2% bovine
serum albumin (w/v)} was incubated for 4 hr at 37C. The
reaction with p-nitrophenyl phosphate (Sigma) (1.0 mg/ml in
10% triethanolamine, pH 9.8) was quantified after 60 min at

63
405 nm (OD405) using a Labinstruments model EAR 400 AT ELISA
plate spectrophotometer.
DAS-indirect ELISA.- For DAS-Indirect ELISA the plates
were coated with polyclonal IgG No. 1053, and antigen samples
were added as described for DAS-ELISA. The 3DF1 MCA was used
at a concentration of 0.2 /g/ml IgG and MCA-13 ascites fluid
was used at a dilution of 1:5,000 (v/v). Each was diluted in
conjugate buffer, added to the appropriate plates and
incubated for 4 hr 37C. After washing, alkaline phosphatase
conjugated goat anti-mouse IgG was added at a dilution of
1:7,500 (v/v) in conjugate buffer and incubated for 2 hr at
37C. The enzyme-substrate reaction was carried out as
described for DAS-ELISA.
Evaluation.- Twelve selected naturally-occurring Florida
CTV isolates with different biological properties (Garnsey et
al. 1987; Permar et al. 1990; Rocha-Pea and Lee, unpublished)
were used to determine the range of reactivity of each
antibody. All virus source plants were maintained in a
greenhouse with mean minimum and maximum temperatures of 21
and 38C, respectively. The isolates Tila, T26, T30, T50a and
T55a produce very mild symptoms and little or no stunting in
Mexican lime seedlings {Citrus aurantifolia (Christm.)
Swingle}, no seedling yellows on sour orange (C. aurantium L.)
or grapefruit (C. paradisi Macf.) and no symptoms in sweet
orange {C. sinensis (L.) Osb.} and sweet/sour orange
combinations. The isolate T4 causes a moderate or stronger

64
reaction on Mexican lime than isolates listed above. The
isolates T3, T36, T62a, T65a, T66a and T67a cause strong vein
clearing, stunting and stem pitting in Mexican lime seedlings,
varying degrees of decline and/or growth reduction in
sweet/sour orange combinations, varying degrees of seedling
yellow reaction and/or severe stunting in sour orange or
grapefruit seedlings, and varying degrees of growth reduction
in Madam Vinous sweet orange seedlings.
The relative sensitivity limits were compared for
detection of CTV by DIBA, DAS-ELISA, and DAS-indirect ELISA.
An extract {1:10 (w/v) in TBS-Tween} was made of greenhouse
grown C. excelsa plants infected with the CTV isolate T36.
A series of two fold dilutions was made with a similar extract
of healthy C. excelsa. Negative controls included TBS-Tween,
and 1:10 extracts of healthy C. excelsa. Madam Vinous sweet
orange, Duncan grapefruit and Mexican lime plants, all in TBS-
Tween.
Evaluation of different buffers for sample extraction.-
An additional experiment was carried out to determine the
effects of different extraction buffers on the sensitivity of
DIBA, DAS-ELISA, and DAS-indirect ELISA. Three different
extraction solutions, Tris buffered saline, phosphate buffered
saline, and carbonate buffer (all as described above), with
and without 0.05% Tween 20, were used. Three CTV isolates
T26, T62a, and T66a propagated in Madam Vinous sweet orange
plants were tested in DIBA, DAS-ELISA, and DAS-indirect ELISA

65
yf-
with polyclonal IgG No. 1053 and the 3DF1 MCA as described
before.-^ Polyclonal IgG No. 1053 (1.0 /Ltg/ml) was incubated for
2 hr at 37C in 1% BSA, 2% PVP, TBS containing 1:200 (v/v)
buffer extract of bark from healthy Madam Vinous sweet orange,
prior to use in DIBA. Substrate reaction for DAS-ELISA and
DAS-indirect ELISA was quantified after 60 min.
Results
Development of the dot-immunobindinq assay. The
reactivity of each antibody and level of background on the
nitrocellulose membranes varied with each IgG used, IgG
concentration and incubation time. Reactions were affected
by the blocking solutions and buffers used for dilution of
the antibodies and the commercial goat anti-species IgG
conjugates.
Six different solutions were evaluated as blocking
agents. TBS alone (Fig. 4.1 A) and 10% horse serum (not
shown) did not prevent the nitrocellulose sheets from turning
dark; whereas 3% BSA, 3% gelatin, 0.5% non-fat dry milk, and
5% Triton X-100 all gave an acceptably white membrane with the
different polyclonal IgGs tested (Fig 4.1 B, C, D, E) The
3% gelatin blocking solution gave the best contrast between
the green color for healthy samples and different intensities
of a purple color for CTV infected samples (Fig. 4.1 C). The
use of Triton X-100 as a blocking agent partially removed the
green material from the nitrocellulose sheets; however, a

66
Fig. 4.1 Effect of different blocking solutions on the
reaction of polyclonal antibodies no. 1053 in dot-
immunobinding assay with citrus tristeza virus (CTV) isolates:
A. TBS alone; B. 3% bovine serum albumin (BSA); C. 3% gelatin;
D. 0.5% non-fat dry milk; E. 5% Triton X-100. Number 1, CTV
T66a; 2, CTV T26; 3, buffer extract from healthy sweet orange
plants; 4, TBS-Tween. Reaction conditions are described in
materials and methods.

67
slight pink reaction was consistently obtained with sap from
healthy plants (Fig. 4.1 E).
Polyclonal IgGs had to be cross absorbed with buffer
extracts of healthy tissue to reliably discriminate between
healthy and CTV infected samples (Fig. 4.2 A, B) and
reactivity varied with concentration and incubation times.
The strongest reactions and best discrimination between CTV
positive and negative samples were achieved using 1.0 ng
IgG/ml and 30 min incubation. An incubation time of 60 min
with 1.0 or 0.1 ng IgG/ml, or longer (18 hr) even with as low
as 0.01 jug IgG/ml frequently resulted in the occurrence of
nonspecific reactions with healthy samples, even when the IgG
had been cross absorbed with extracts of healthy plants. sA
would be expected, MCAs did not have to be cross absorbed with
buffer extracts from healthy plants prior to use. The
reactivity of the 3DF1 and MCA-13 MCAs was substantially lower
than that obtained with any of the polyclonal IgGs tested.
For both 3DF1 (1.0 /xg/ml) and MCA-13 (diluted 1:5,000), the
incubation time was extended to 2 hr or longer (18 hr) to
achieve an adequate positive reaction with infected samples
(Fig. 4.2 C, D). The reactivity of 3DF1 was always slightly
lower than that obtained with either polyclonal IgG or the
MCA-13.
In initial experiments with polyclonal IgGs, 1% BSA in
TBS plus 2% PVP was chosen as the antibody diluent because it
consistently gave a whiter background on the nitrocellulose

68
A
B
C
D
i


2 t
;
3
4
Fig. 4.2 Reaction of polyclonal and monoclonal antibodies in
dot-immunobinding assay with citrus tristeza virus (CTV)
isolates. A. Polyclonal antibodies 1053 (1.0 /g/ml) not
cross-absorbed with buffer extract of healthy plant. B.
Polyclonal antibodies 1053 (1.0 /g/ml) incubated with 1:200
(v/v) buffer extract from healthy plants for 2 hr at 37C
prior to use. C and D, monoclonal 3DF1 (1.0 /ig/ml) and MCA-
13 (1:5,000 dilution) antibodies not cross-absorbed with
healthy extract. Number 1, CTV T66a; 2, CTV T26; 3, buffer
extract from healthy sweet orange plants; 4, TBS-Tween.
Reaction conditions are described in materials and methods.

69
and stronger reactivity for positive samples. Either the
absence of PVP or BSA concentrations of less than 1% in the
antibody diluent frequently resulted in a dark background on
the nitrocellulose when 3% BSA was used for blocking.
Evaluation.- The relative sensitivity level of DIBA was
measured using increasing dilutions of a plant extract
containing CTV isolate T36. This is shown in Figure 4.3.
Polyclonal IgGs Nos. 1051, 1052, and 1053 (1.0 xg/m 1 IgG for
30 min) consistently gave a distinct positive reaction with
CTV T36 diluted to 1/160. There was a slight but consistently
positive reaction at the 1/320 dilution and occasionally at
1/640 (Fig. 4.3 A, B, C) The reactivity of both 3DF1 and
MCA-13 MCAs (1.0 /g/ml and 1:5,000, respectively) was also in
the range of 1/160 and 1/320, but only when the incubation
time was extended to 18 hr (Fig. 4.3 D, E).
The relative sensitivities of DAS-ELISA and DAS-indirect
ELISA were measured similarly against the diluted extract of
CTV T36. These results are summarized in Table 4.1. DAS-
ELISA, using polyclonal IgG No. 1053 for both coating and
conjugate steps, gave OD405 value of 0.109 (i.e above 0.100)
with the CTV T36 sample diluted to 1/160. This was considered
a positive reaction. Thus the OD405 value of 0.070 at a 1/320
dilution was considered a negative reaction. DAS-indirect
ELISA, using polyclonal IgG No. 1053 for coating and 3DF1 MCA
as second antibody in the double sandwich, gave a positive and
negative reaction at dilutions of 1/320 and 1/640,

70
1/10
1/20
1/4 0
1/80
1/160
1/320
1/640
TBS-Tween
Fig. 4.3 Relative sensitivity level of different polyclonal
and monoclonal antibodies specific to citrus tristeza virus
(CTV) in dot-immunobinding assay. Rows A, B and C, polyclonal
antibodies nos. 1051, 1052, and 1053, respectively. Rows D
and E, monoclonal antibodies 3DF1 and MCA-13, respectively.
Extract of Citrus excelsa greenhouse grown plants infected
with CTV T36 isolate was prepared in TBS-Tween 1:10 (w/v) and
two-fold diluted with extract of healthy plants. Reaction
conditions are described in materials and methods.

Table 4.1 Relative
polyclonal (1053) and
tristeza virus.
sensitivity
monoclonal
Of DIBA,
(3DF1 and
DAS-ELISA and DAS-
MCA-13) antibodies
indirect
specific
ELISA with
to citrus
CTV T-3 6
isolate'
DIBA
DAS-ELISA1/
DAS-indirect
ELISA1/
1053
3DF1
MCA-13
1053
3DF1
MCA-13
1/10
' +1/
+
+
0.572
+4/
2.628 +
0.892 +
1/20
+
+
+
0.460
+
2.507 +
0.607 +
1/40
+
+
+
0.314
+
1.635 +
0.385 +
1/80
+
+
+
0.185
+
0.853 +
0.252 +
1/160
+
+
+
0.109
+
0.518 +
0.158 +
1/320
+
+/-
+/-
0.070
-
0.190 +
0.094 -
1/640
-
-
-
0.024
-
0.075 -
0.054 -
1/1280
-
-
-
0.022
-
0.032 -
0.036 -
C. excelsa
-

-
0.004
-
0.033 -
0.013 -
The IgG of polyclonal antibody no. 1053 was used to coat the plates for both DAS-
ELISA and DAS-indirect ELISA. For DAS-ELISA the no. 1053 IgG conjugate was used as
second antibody. For DAS-indirect ELISA the unlabeled 3DF1 and MCA-13 were the
intermediate antibodies followed by the goat anti-mouse IgG conjugate.
1/

2/
Buffer extract (1:10 dilution) of bark from greenhouse grown C. excelsa plants
infected with CTV T36 successively two-fold diluted with healthy C. excelsa buffer
extract.
3 /
' Visual evaluation for presence of a purple color (+ = positive, +/- = inconclusive, -
= negative.
Optical density at 405 nm (OD4Q5) after 60 min of substrate reaction. Mean of two
replicatipns per plate. Reactions were considered positive (+) when OD4Q5 values were
higher than three times the mean of healthy controls or 0.100, whichever was greater.
Reactions with lower values were considered negative (-).

73
respectively. DAS-indirect ELISA using the MCA-13 as second
antibody had a sensitivity level similar to DAS-ELISA i.e.
the dilution end point was 1/160 dilution (Table 4.1).
The relative reactivity of the different antibodies for
the 12 selected CTV isolates in DIBA, and its comparison with
both DAS-ELISA and DAS-indirect ELISA, is illustrated in
Figure 4.4 and Table 4.2. Polyclonal IgG No. 1053 (1.0 /ng/ml)
showed a strong positive reaction with all CTV isolates tested
in DIBA (Fig. 4.4, left). Polyclonal IgGs Nos. 1051 and 1052
reacted similarly (not shown). The results obtained with
polyclonal IgG No.1053 in DAS-ELISA with the CTV isolates were
the same as with DIBA (Table 4.2).
The 3DF1 MCA (1.0 /xg/ml IgG) reacted moderately with most
CTV isolates tested, but no reaction and weak or inconclusive
reactions were obtained with the isolates T26 and T66a,
respectively, in both DIBA and DAS-indirect ELISA (Fig. 4.4,
center and Table 4.2).
The severe strain specific MCA-13 (used at a dilution of
1:5,000) gave a distinctly positive reaction only with
isolates T3, T36, T65a, T66a, and T67a in DIBA and DAS-
indirect ELISA (Fig. 4.4, right and Table 4.2). An
inconclusive or slightly positive reaction was obtained
occasionally with the T50a, T55a, T4, and T62a isolates in
DIBA when the samples were incubated for longer times (18 hr) .
Similar reactions occurred with these particular CTV isolates
in the DAS-indirect ELISA test (Table 4.2).

74
1053 3DF1 MCA-13
Fig. 4.4 Reaction of polyclonal antibodies no. 1053 and 3DF1
and MCA-13 monoclonal antibodies in dot-immunobinding assay
with twelve selected citrus tristeza virus (CTV) isolates.
Row A: 1 = Tila; 2 = T26; 3 = T30; 4 = T50a; 5 = T55a; 6 = T3;
7 = T4; 8 = T36. Row B: 1 = T62a; 2 = T65a; 3 = T66a; 4 =
T67a. Also in Row B are extract of healthy plants: 5 = Citrus
excelsa; 6 = Madam Vinous sweet orange; 7 = Duncan grapefruit;
8 = Mexican lime. Reaction conditions are described in
materials and methods.

Table 4.2 Comparison of DIBA, DAS-ELISA and DAS-indirect ELISA for the detection of
citrus tristeza virus (CTV) and relative reactivity of polyclonal (1053) and monoclonal
(3DF1 and MCA-13) antibodies with CTV isolates.
Virus
DIBA
isolate'
1053
3DF1
MCA-13
Tila
(M)1/
+1/
+
-
T2 6
(M)
+
+/-
-
T30
(M)
+
+
-
T50a
(M)
+
+
+/-
T55a
(M)
+
+
+/-
T3
(S)
+
+
+
T4
(HD)
+
+
+/-
T36
(S)
+
+
+
T62a
(S)
+
+
+/-
T65a
(S)
+
+
+
T66a
(S)
+
+/-
+
T67a
(S)
+
+
+
DAS-ELISA1^ DAS-indirect ELISA1/
1053
3DF1
MCA-13
0.204
+5/
1.568
+
0.091
0.137
+
0.053
-
0.057
0.227
+
1.936
+
0.090
0.498
+
2.584
+
0.120
0.443
+
2.018
+
0.119
0.404
+
2.468
+
0.522
0.304
+
1.992
+
0.092
0.572
+
2.628
+
0.892
0.556
+
2.045
+
0.115
0.556
+
2.527
+
0.745
0.588
+
0.191
+
0.735
0.250
+
2.235
+
0.480
ui

C. excelsa
(HC) -


0.021 -
0.033 -
0.013
M. Vinous
(HC) -
-
-
0.004 -
0.045 -
0.000
Grapefruit
(HC) -
-
-
0.004 -
0.014 -
0.008
M. lime
(HC) -
-
-
0.009 -
0.029 -
0.038
J The IgG of polyclonal antibody no. 1053 was used to coat the plates for both DAS-
ELISA and DAS-indirect ELISA. For DAS-ELISA the no. 1053 IgG conjugate was used as
second antibody. For DAS-indirect ELISA the unlabeled 3DF1 and MCA-13 were the
intermediate antibodies followed by the goat anti-mouse IgG conjugate.
2 / ,
' Buffer extracts (1:10 dilution) of bark from greenhouse grown plants either of C.
excelsa. Madam Vinous sweet orange, grapefruit or Mexican lime.
3 / ,
' M = mild; MD = moderate; S = severe; HC = healthy control; on the basis of symptom
reaction on a series of citrus hosts (Garnsey et al. 1987; Permar et al. 1990; Rocha-
Pea and Lee, unpublished).
/ Visual evaluation for presence of a purple color (+ = positive, +/- = inconclusive, -
= negative).
Optical density at 405 nm (D405) after 60 min of substrate reaction. Mean of two
replications per plate. Reactions were considered positive (+) when OD4Q5 values were
higher than three times the mean of healthy controls or 0.100, whichever was greater.
Reactions with lower values were considered negative (-).

77
Evaluation of different buffers for sample extraction.-
The reactions of the polyclonal IgG 1053 and MCA 3DF1 with
the different CTV isolate/extraction buffer combinations in
DIBA are shown in Figure 4.5. In general, there was a strong
positive reaction with all the CTV isolate/extraction buffer
combinations tested, with the strongest reaction occurring
when PBS without Tween 20 was used as the extraction buffer
(Fig. 4.5). The presence of Tween 20 in the extraction buffer
gave a weaker but relatively more uniform reaction spot with
both antibodies.
The results of the evaluation of the different extraction
buffers by DAS-ELISA and DAS-indirect ELISA are shown in
Figure 4.6. In DAS-ELISA the strongest reactions occurred
with isolate T62a, followed by T26 and T66a, respectively.
The presence or absence of Tween 20 made little difference in
DAS-ELISA (Fig. 4.6 left). In DAS-indirect ELISA with the
3DF1 MCA the strongest reactions occurred with isolate T26,
followed by T62a and T66a, respectively. The presence of
Tween 20 gave slightly stronger reactions in most cases. With
PBS, the presence of Tween 2 0 more than doubled the OD405
values for isolates T62a and T66a, but caused only a slight
increase for isolate T26 (Fig. 4.6 right)
Discussion
The dot-immunobinding assay was adapted for detection of
CTV. This included the testing of different agents for
blocking and as diluents for the CTV specific antibodies and

78
si I I
4>
*-
a
TBS
4


TBST



PBS

o

PBST


%
Carb



CarbT
O


H20



o
O
1053
3DF1
Figure 4.5 Evaluation of different extraction buffers on the
sensitivity of DIBA with citrus tristeza virus (CTV) isolates
T26, T62a and T66a. TBS = Tris buffered saline, TBST = TBS
containing 0.05% Tween 20, PBS = phosphate buffered saline,
PBST = PBS + 0.05% Tween, Carb = carbonate buffer, CarbT =
Carb + 0.05% Tween, HC = healthy control. The IgG of
polyclonal antibodies No. 1053 or monoclonal antibody 3DF1
were used as primary antibodies followed by the goat anti
rabbit and anti-mouse IgG conjugates, respectively. Samples
were 2 /I of buffer extracts of greenhouse-grown Madam Vinous
sweet orange plants infected with the CTV isolates ground at
a 1:10 dilution.

Optical Density
DAS-ELISA
DAS-indirect ELISA
Figure 4.6 Evaluation of different extraction buffers on the sensitivity of DAS-ELISA
and DAS-indirect ELISA with citrus tristeza virus (CTV) isolates T26, T62a and T66a.
TBS = Tris buffered saline, TBST = TBS containing 0.05% Tween 20, PBS = phosphate
buffered saline, PBST = PBS + 0.05% Tween, Carb = carbonate buffer, CarbT = Carb + 0.05%
Tween, HC = healthy control. The IgG of polyclonal antibodies No. 1053 was used to coat
the plates for both DAS-ELISA and DAS-indirect ELISA. For DAS-ELISA the No. 1053 IgG
conjugate was used as second antibody. For DAS-indirect ELISA the unlabeled 3DF1
monoclonal antibody was the intermediate antibody followed by the goat anti-mouse IgG
conjugate. Samples were 200 /I of buffer extracts of greenhouse-grown Madam Vinous
sweet orange plants infected with the CTV isolates ground at a 1:10 dilution.

80
the commercial goat anti-species IgG conjugates, the
evaluation of some CTV specific polyclonal and monoclonal
antibodies and the testing of the effect of different
extraction buffers with and without Tween 20.
In some other systems where nitrocellulose membranes have
been used for the immunological detection of proteins either
by DIBA (Hibi and Saito, 1985; Powell, 1987; Graddon and
Randles, 1986) or by western blotting (Spinola and Cannon,
1985), some differences have been reported in the suitability
of different agents and buffers used for membrane blocking or
as diluents for both the virus specific antibodies and the
commercial goat anti-species IgG conjugates. In this study,
TBS, 10% horse serum, 3% BSA, 3% gelatin, 0.5% non-fat dry
milk and 5% Triton X-100 were evaluated as blocking agents.
The latter four were found to give an adequately white
background on the nitrocellulose membranes to permit
discrimination between infected and healthy samples. However,
3% gelatin was used routinely as it gave the most suitable
contrast between a green color for the healthy samples and a
distinct purple color for the infected samples.
It has been well documented that the presence of albumin
(BSA or egg ovalbumin) enhances the reactivity of antibodies
in diverse serological tests (Clark et al. 1986; Purcifull and
Batchelor, 1977). Likewise, PVP has been commonly used in
both extraction and conjugate buffers to prevent non-specific
reactions (Clark et al. 1986). In this study, a concentration

81
of BSA of less than 1% or the absence of PVP in the antibody
diluent frequently resulted in the nitrocellulose membrane
turning dark when BSA was used for blocking. Thus TBS
containing 1% BSA and 2% PVP was always used as the antibody
diluent.
Differences were found in the reactivities of the
antibodies tested. All polyclonal IgGs (Nos. 1051, 1052, and
1053) reacted similarly in DIBA. After 30 min incubation (at
1.0 jul/ml IgG) they gave strong positive reactions with CTV-
infected samples, but none with healthy samples. However,
incubations of 60 min or more sometimes resulted in the
occurrence of nonspecific reactions with extracts of healthy
plants. This was prevented when the polyclonal antibodies
were cross-absorbed with extracts of healthy plants before
use. These antibodies gave a strong positive reaction with
the 12 selected CTV isolates tested, and had a sensitivity
(highest dilution which gave positive reaction) limit of 1/320
with CTV T36 infected samples.
The 3DF1 MCA when used at an IgG concentration of 1.0
/xg/ml reacted moderately with most of the CTV isolates tested
and had a relative sensitivity limit of 1/320 in DIBA and
1/680 in DAS-indirect ELISA. However, in some cases it was
necessary to extend the incubation time (15-18 hr) with 3DF1
in DIBA, because it had a reactivity level that was somewhat
lower than any of the polyclonal antibodies tested.

82
The 3DF1 MCA has been reported to react with a broad
spectrum of CTV isolates primarily on the basis of DAS-
indirect ELISA (Vela et al. 1986, 1988) In this work the
3DF1 reacted weakly with CTV isolates T26 and T66a in some
tests (Fig. 4.4, center, A2 and B3) and moderately in others
(Fig. 4.2, Cl,2,5,6). The reason for this is not known. For
isolate T26 it could be due to a low virus concentration as
was indicated by DAS-ELISA (Table 4.2). However, the T66a
isolate was at a high concentration as indicated in DAS-ELISA
with polyclonal antibody 1053 and in DAS-indirect ELISA with
MCA-13 (Table 4.2). Yet, in DAS-indirect ELISA 3DF1 gave a
low OD405 value for T66a (Table 4.2). This may indicate a
differential reactivity of the 3DF1 MCA with some particular
CTV isolates such as T66a.
The sensitivity limit for MCA-13 in both DIBA and DAS-
indirect ELISA was near 1/320 (Fig. 4.3 and Table 4.1). In
DIBA MCA-13 reacted strongly with CTV isolates T3, T36, T65a,
T66a, and T67a, all of which share severe biological
properties (Garnsey et al. 1987; Permar, et al. 1990; Rocha-
Pea and Lee, unpublished). However, there were slight
positive reaction with CTV isolates T50a and T55a which have
mild biological properties and with isolate T62a which has
severe properties (Rocha-Pea and Lee, unpublished) (Table
4.2) The MCA-13 has been reported to react specifically with
CTV isolates that have severe biological properties, such as
decline, stem pitting, seedling yellows, etc. (Permar and

83
Garnsey, 1990; Permar et al. 1990), and it has been used for
strain discrimination of CTV-induced decline in several field
surveys (Irey, et al. 1988; Rocha-Pea et al. 1990; Yokomi et
al. 1990). The slight positive reaction found frequently in
DIBA with some CTV isolates with mild biological properties
occurred when the ascites fluid at a dilution of 1:5,000 was
incubated for 18 hr. These slight positive reactions could
be prevented by using a shorter (2 hr) incubation time or by
using the ascites fluid at a dilution of 1:20,000 and an
incubation time of 18 hr (Rocha-Pea et al. 1990) The
isolate T62a has been characterized previously as severe on
the basis of a severe vein flecking, leaf cupping and stunting
of Mexican lime plants, severe growth reduction on grapefruit
and Madam Vinous sweet orange seedlings, and moderate growth
reduction on sweet/orange combinations (Rocha-Pea and Lee,
unpublished). The weak reaction found with isolate T62a in
DIBA and its low OD405 when MCA-13 was used in DAS-indirect
ELISA (Table 4.2), suggests that some severe CTV isolates may
not be recognized by the MCA-13. That T62a was in high
concentration in these experiments is verified by DAS-ELISA
using polyclonal antibodies and DAS-indirect ELISA using the
3DF1 MCA (Table 4.2).
Three buffers, TBS, PBS and carbonate, with and without
0.05% Tween 20, were evaluated for their effects on the
sensitivity of DIBA, DAS-ELISA and DAS-indirect ELISA with
two antibodies and three CTV isolates. In DIBA, PBS gave the

84
strongest reactions, followed by TBS and carbonate. The
addition of Tween 20 gave slightly weaker reactions with all
buffers, but the spots were more uniformly spread on the
nitrocellulose. With both ELISA procedures the presence of
Tween 20 in the extraction buffers tended to increase the
OD405 values slightly. However, in DAS-indirect ELISA, the
presence of Tween 20 in the PBST more than doubled OD405
readings over PBS alone for CTV isolates T62a and T66a, but
had little effect with isolate T26. This did not occur with
TBS or the carbonate buffer. One possible explanation is that
the presence of Tween 20 in the PBST caused a conformational
change in the coat protein of T62a and T66a exposing more of
the epitope for binding with the 3DF1 MCA. This appeared to
occur to a much lower degree with isolate T26 in DAS-indirect
ELISA and with all three of these isolates reacting with the
polyclonal antibodies in DAS-ELISA. These results indicate
that PBS might be the best extraction buffer for DIBA, and
PBST or TBST might be the best for ELISA. However, it would
always be advisable to determine the effects of Tween 20 on
the serological methods and antigen/antibody combinations
under investigation.
There are several advantages to using DIBA over
conventional DAS-ELISA or DAS-indirect ELISA for CTV
detection. DIBA was rapid and easy to perform and, it was as
sensitive as either ELISA procedure for CTV diagnosis. The
entire test could be performed in 2-3 hours using polyclonal

85
antibodies (slightly longer with monoclonal antibodies), and
minimal laboratory equipment was needed. The use of small
containers such as the polypropylene covers for incubation and
washing steps enabled the recovery and re-use of both virus
specific antibodies and goat anti-species IgG conjugates for
at least four weeks when the solutions were properly preserved
with 0.02% sodium azide and stored at 5C between uses. The
template constructed from the rack of a micropipet tip holder
box enabled the uniform spacing of samples on the
nitrocellulose membranes.
As pointed out by Powell (1987) one disadvantage of DIBA
as compared to ELISA procedures is the lack of quantitative
measurements. However, with appropriate positive and negative
controls, DIBA can be used reliably for routine diagnostic
work where no quantitative measurements are needed. When
quantitation of CTV antigens is required, ELISA should be
used.

CHAPTER 5
SUMMARY AND CONCLUSIONS
Four different naturally occurring Florida mild citrus
tristeza virus (CTV) isolates were evaluated under greenhouse
conditions at two temperature regimes for their cross-
protecting ability against the induced decline by a CTV severe
challenge isolate in two susceptible scion/rootstock
combinations. DAS-ELISA with polyclonal antisera was used to
determine the total antigen titer in plants inoculated with
mild isolates and those challenged with the severe isolate.
The MCA-13 strain specific monoclonal antibody was
successfully used in DAS-indirect ELISA for differential
isolate detection and quantitation of the severe challenge
isolate in mixed infections.
The effectiveness of the graft transmission of CTV was
evaluated with three CTV isolates by using leaf and bark
tissue from three citrus donor hosts to three receptor hosts.
The distribution of the virus in different plant tissues was
also studied.
A dot-immunobinding assay (DIBA) was adapted for CTV
detection. The sensitivity level of DIBA was evaluated using
three different polyclonal and two monoclonal antibodies and
86

87
compared with sensitivities of double antibody sandwich (DAS)
ELISA and DAS-indirect ELISA with 12 different CTV isolates.
The conclusions of this research are summarized as follows:
Cross-Protection of Citrus Tristeza Virus
At warm temperatures plants pre-inoculated with mild
isolates showed relatively low optical density (OD405) values
in the range of 0.091-0.145 in DAS-ELISA with polyclonal
antisera; whereas, at cool temperatures the values were
commonly in the range of 0.300-0.400. At warm temperatures,
the MCA-13 monoclonal antibody in plants pre-inoculated with
mild isolates but unchallenged, gave OD405 values in the range
of 0.030-0.045 which were as low as the uninoculated control
plants. However, at cool temperatures, the OD405 values
obtained with mild isolates were slightly higher than those
obtained at warm temperatures, indicating that the MCA-13
monoclonal antibody may react to some extent with mild
isolates when they are above a certain titer in the plants.
Nevertheless, the OD405 values were always lower than 0.100,
which was considered a negative reaction.
When plants pre-inoculated with mild isolates and further
challenged with the severe isolate were assayed with the MCA-
13 monoclonal antibody, they gave typically lower OD405 values
for the T66a isolate than the unprotected challenged control
plants which did not have mild isolates. In this regard, the
T26 and T3 0 isolates gave the more uniform lower OD405 values
for the T66a isolate in all evaluated treatments. This would

88
indicate that the mild isolates, especially the T26, T30, and
T55a, prevented to some extent the multiplication of the
challenge severe isolate, and that they were apparently
working as cross-protecting agents. The evaluation of the
virus titer of the T66a severe isolate with the MCA-13
monoclonal antibody in the treatments pre-inoculated with mild
isolates and further challenged provided a measurable
parameter to estimate the ability of mild isolates to prevent
the establishment the severe isolate.
The effect of mild CTV isolates on performance of both
Valencia/sour orange and Valencia/macrophylla was variable.
At warm temperatures, the effect was negligible. The decline
index for the healthy uninoculated control plants was nearly
equal to or greater than those inoculated only with mild
isolates. However, at cool temperatures there was a
remarkable growth reduction effect in all treatments
evaluated. The decline indexes for the healthy uninoculated
control plants was variable and ranged from values below to
above those obtained with plants inoculated only with mild
isolates. Of all treatments evaluated at both temperatures,
the T26 isolate, and also T55a to some degree, obtained the
lowest decline index scores and lowest number of dead plants
as compared with the uninoculated challenged control plants.
This provided further evidence of their cross-protecting
effect, especially the T26 isolate against the development of
the CTV-ID syndrome.

89
Plants pre-inoculated with the Tila mild isolate, and
further challenged with the T66a severe isolate showed the
highest occurrence of decline, reflected in high decline index
scores in the whole experiment, indicating apparently that the
combination of both Tila and T66a isolates produced a more
severe reaction in the challenged plants even than that caused
by the T66a isolate alone in the unprotected control plants.
There was a relatively low occurrence of decline at both
temperatures in the unprotected control which were challenged,
indicating that to better guarantee an appropriate occurrence
of decline symptoms under greenhouse conditions, it would be
advisable toe use a mixture of more than one severe isolate
as the challenge.
The results of this work provide further evidence that:
a) the cross-protection against the induced decline on
sweet/sour orange combinations is possible, and it can be
evaluated preliminarily under greenhouse conditions in a
relatively short time period of 18-24 months. This time period
estimate includes from six to twelve months to infect the test
plants with mild isolates and verification of infection by
serology, two months for challenge inoculation with the severe
isolate and ten months for final evaluation; b) The severe
challenge isolate can be differentially detected from the mild
isolates by using the MCA-13 strain specific monoclonal
antibody; c) Mild isolates can be evaluated without the risk

90
of recurrent freezes that can hamper the reliable evaluation
of cross-protection experiments under field conditions.
Graft Transmission of Citrus Tristeza Virus
The efficiency of graft transmission of CTV by leaf or
bark pieces was conditioned primarily by the donor/receptor
host combination and secondly by the virus isolate involved.
For example, C. excelsa showed a low 72.4% and 60.7% rate of
transmission to Madam Vinous and grapefruit, respectively;
whereas, an 86.7% rate of transmission was obtained to Mexican
lime plants. In regard to Mexican lime as donor hosts, there
was a rate of transmission of 89.3% and 93.1.7% to grapefruit
and Madam Vinous, respectively and a 76.9% rate to Mexican
lime. Rate of transmission from Madam Vinous sweet orange was
between 84.6 and 100% in all receptors tested. Of the three
donor hosts tested, Madam Vinous sweet orange was the most
efficient donor host (90.6%), followed by Mexican lime
(85.2%). C. excelsa was a poor donor host with an overall
rate of transmission of 72.5%, being relatively efficient only
when inoculated to Mexican lime. Differences were found
in the rate of transmission of the three different CTV
isolates with each donor host tested. However, the overall
rate of transmission was in the range of 80% for all isolates.
Apparently the efficiency of the graft transmission of the
virus depended more on the donor host involved than the virus
isolate.

91
There were some differences in the virus concentration
in the donor tissues used as inoculum. Bark tissue contained
the highest titer with OD405 values in the range of 0.221 and
0.349 in both Madam Vinous and C. excelsa with both CTV
isolates tested. These values were, in some instances, more
than double those found in petioles and midribs, and at least
triple those found in the leaf blade. While these values were
not statistically significant, some differences were found in
virus titer from different parts of the same plant and from
one plant to another. There were some instances where OD405
values were as low as the healthy controls indicating a
possible absence of the virus in those tissues. This raises
the possibility that occasionally the tissue used for graft
transmission may be virus-free, with a subsequent failure in
the transmission.
The use of leaf and/or bark pieces for graft transmission
of CTV may be advantageous when large numbers of plants are
to be inoculated with limited sources of inoculum, as is the
case for studies of cross-protection. However, in order to
achieve a high level in the rate of transmission, the
efficiency of donor host and the donor/receptor host
combination should be considered. Leaf pieces provide a
better source for inoculation than bark tissue.
A Dot-Immunobindinq Assay for Citrus Tristeza Virus
The dot-immunobinding assay (DIBA) was adapted for
detection of CTV. Some differences were found in the
J

92
reactivity of each antibody with the different CTV isolates
tested. All polyclonal antibodies (nos. 1051, 1052, and 1053)
-
reacted similarly in DIBA, but all had to be cross-absorbed
with buffer extracts of healthy plants before use. All
polyclonals gave a strong positive reaction with the 12 CTV
isolates tested. The dilution end points of extracts from
CTV-infected samples were 1/320 when tested against the
polyclonal antibodies.
The 3DF1 monoclonal antibody, at a concentration of 1.0
ig/ml,reacted moderately with most of the CTV isolates tested.
Dilution end points of extracts from CTV-infected tissue were
1/320 in DIBA and 1/680 in DAS-indirect ELISA. The strongest
reactions in DIBA were obtained when the incubation time was
extended to 18 hr.
The MCA-13 monoclonal antibody used, at a dilution of
1:5,000 in DIBA, reacted strongly with CTV isolates T3, T36,
T65a, T66a, and T67a all sharing common severe biological
properties. However, there was a slight positive reaction
with two CTV isolates, T50a and T55a, having mild biological
properties and with one isolate (T62a) known to have severe
properties. The highest dilution end point of CTV-infected
extracts which reacted with MCA-13 in DIBA and DAS-indirect
ELISA was 1/320 using an incubation time of 18 hr.
There are several advantages to using DIBA over
conventional DAS-ELISA or DAS-indirect ELISA for CTV
detection. DIBA was rapid and easy to perform and, it was as

93
sensitive as either ELISA procedure evaluated for CTV
diagnosis. The entire test could be performed in 2-3 hours
using polyclonal antibodies (slightly longer with monoclonal
antibodies), and minimal laboratory equipment was needed. The
use of small containers such as the polypropylene covers for
incubation and washing steps enabled the recovery and re-use
of both virus specific antibodies and goat anti-species IgG
conjugates for at least four weeks when the solutions were
properly preserved with 0.02% sodium azide and stored at 5C.
The template constructed from the rack of a micropipet tip
holder box enabled the uniform spacing of samples on the
nitrocellulose membranes.
One disadvantage of DIBA as compared to ELISA procedures
is the lack of quantitative measurements. However, with
appropriate positive and negative controls, DIBA can be used
reliably for routine diagnostic work where no quantitative
measurements are needed. When quantitation of CTV antigens
is required, ELISA should be used.

LITERATURE CITED
Balaraman, K. and Ramakrishnan, K. 1980. Strain variation and
cross protection in citrus tristeza virus. Pages 60-68,
in: Calavan, E.C., Garnsey, S.M., and Timmer, L.W. (eds) .
Proc. 8th Conf. Intern. Organ. Citrus Virol. Riverside,
California.
Bar-Joseph, M. and Lee, R.F. 1990. Citrus tristeza virus.
Description of Plant Viruses No. 353 (No. 33 revised).
Commonwealth Mycological Institute/Association of Applied
Biologists. Kew Surrey, England. 7 p.
Bar-Joseph, M. and Loebenstein, G. 1973. Effects of strain,
source plant, and temperature on the transmissibility of
citrus tristeza virus by the melon aphid. Phytopathology
63:716-720.
Bar-Joseph, M. and Malkinson, M. 1980. Hen yolk as a source
of antiviral antibodies in the enzyme-linked
immunosorbent assay (ELISA): a comparison of two plant
viruses. J. Virol. Meth. 1:1-5.
Bar-Joseph, M., Garnsey, S.M., and Gonsalves, D. 1979a. The
closteroviruses: a distinct group of elongated plant
viruses. Adv. Virus Res. 25:93-168.
Bar-Joseph, M., Garnsey, S.M., Gonsalves, D., and Purcifull,
D.E. 1980. Detection of citrus tristeza virus. I. Enzyme-
linked immunosorbent assay (ELISA) and SDS-
immunodiffusion methods. Pages 1-8, in: Calavan, E.C.,
Garnsey, S.M., and Timmer, L.W. (eds). Proc. 8th Conf.
Intern. Organ. Citrus Virol. Riverside, California.
Bar-Joseph, M., Garnsey, S.M., Gonsalves, D., Moscovitz, M.,
Purcifull, D.E., Clark, M.F. and Loebenstein, G. 1979b.
The use of enzyme-linked immunosorbent assay for the
detection of citrus tristeza virus. Phytopathology
69:190-194.
Bar-Joseph, M., Gumpf, D.J., Dodds, J.A., Rosner, J.A., and
Ginzberg, I. 1985. A simple purification method for
citrus tristeza virus and estimation of its genome size.
Phytopathology 75:195-198.
94

95
Bar-Joseph, M., Marcus, R., and Lee, R.F. 1989. The continuous
challenge of citrus tristeza virus control. Ann. Rev.
Phytopathol. 27:292-316.
Bar-Joseph, M., Raccah, B., and Loebenstein, G. 1977.
Evaluation of the main variables that affect citrus
tristeza virus transmission by aphids. Proc. Intern. Soc.
Citriculture 3:958-961.
Bar-Joseph, M., Roistacher, C.N., and Garnsey, S.M. 1983.
Epidemiology and control of citrus tristeza virus. Pages
61-72, in: Plumb, R.T. and Thresh, J.M. (eds) Plant
Virus Epidemiology. Blackwell Sci. Pub. Oxford.
Bar-Joseph, M. Roistacher, C.N., Garnsey, S.M., and Gumpf,
D.J. 1981. A review on tristeza, an ongoing threat to
citriculture. Proc. Intern. Soc. Citriculture 1:419-423.
Beachy, R.N., Abel, P.P., Nelson, R.S., Register, J., Turner,
N.,and Fraley, R.T. 1987. Genetic engineering of plants
for protection against virus diseases. Pages 151-159, in:
Plant Resistance to Viruses. Ciba Foundation Symposium
133. Wiley, Chichester.
Bennett, C.W. and Costa, A.S. 1949. Tristeza disease of
citrus. J. Agr. Res. 78:207-237.
Ben-Ze'ev, I.S., Frank, A., and Bar-Joseph, M. 1988. Sensitive
detection of two plant viruses by enzyme-amplified ELISA.
Phytoparasitica 16:343-349.
Blue, R.L., Roistacher, C.N., G. Cartia, and Calavan, E.C.
1976. Leaf disc grafting: A rapid indexing method for
detection of some citrus viruses. Pages 207-212, in:
Calavan, E.C. (ed.). Proc. 7th Conf. Inter. Organ. Citrus
Virol. Riverside, California.
Brlansky, R.H. 1987. Inclusion bodies produced in Citrus spp.
by citrus tristeza virus. Phytophylactica 19: 211-213.
Brlansky, R.H. 1988. Other bacterial canker diseases. Page 7,
in: Whiteside, J.O., Garnsey, S.M., and Timmer, L.W.
(eds). Compendium of Citrus Diseases. APS Press. 80 p.
Brlansky, R.H., Garnsey, S.M., Lee, R.F., and Purcifull, D.E.
1984. Applications of citrus tristeza virus antisera for
use in labeled antibody, immuno-electron microscopical,
and sodium dodecyl sulphate-immuno-diffusion tests. Pages
337-342, in: Garnsey, S.M., Timmer, L.W., and Dodds, A.J.
(eds). Proc. 9th Conf. Intern. Organ. Citrus Virol.
Riverside, California.

96
Brlansky, R.H., Pelosi, R.R., Garnsey, S.M., Youtsey, C.O.,
Lee, R.F., Yokomi, R.K., and Sonoda, R.M. 1986. Tristeza
quick decline epidemic in South Florida. Proc. Fla. State
Hort. Soc. 99:66-69.
Calavan, E.C., Mather, S.M., and McEachern, E.H. 1978.
Registration, certification, and indexing of citrus
trees. Pages 185-222, in: Reuter, W., Calavan, E.C., and
Carman, E.G. (eds) The Citrus Industry. Vol IV. Crop
Protection. University of California. Riverside,
California.
Clark, M.F. and Adams, A.M. 1977. Characteristics of the
microplate method of enzyme-linked immunosorbent assay
for the detection of plant viruses. J. Gen. Virol.
34:475-483.
Clark, M.F., Lister, R.M., and Bar-Joseph, M. 1986. ELISA
techniques. Methods in Enzymology 118:742-767.
Cohen, M. 1972. A leaf insert graft used for virus
transmission in citrus. Pages 282-284, in: Price, W.C.
(ed) Proc. 5th Conf. Intern. Organ. Citrus Virol.
Gainesville, Florida.
Costa, A.S. and Muller, G.W. 1980. Tristeza control by cross
protection: A U.S.-Brazil cooperative success. Plant
Disease 64:538-531.
Davis, C.L. and Brlansky, R.H. 1991. Rapid detection of citrus
tristeza virus in citrus extracts using a gold labelled
antibody. J. Elect. Microsc. Tech. (Abstr., in press).
DeLange, J.H., Van Vuuren, S.P., and Bredell, G.S. 1981.
Groeipuntenting suiwer sitrusklone vir die
superplantskema van virusse. Subtropica 2 (5):11-16.
Fulton, R.W. 1986. Practices and precautions in the use of
cross protection for plant virus disease control. Ann.
Rev. Phytopathol. 24:67-81.
Garnsey, S.M. and Cambra, M. 1990. Enzyme linked immunosorbent
assays for citrus pathogens. In: Roistacher, C.N. (ed).
F.A.O. Citrus Disease Indexing Bulletin (in press).
Garnsey, S.M. and Jackson, J.L. 1975. A destructive outbreak
of tristeza in central Florida. Proc. Fla. State Hort.
Soc. 88:65-69.

97
Garnsey, S.M. and Lee, R.F. 1988. Tristeza. Pages 48-50, in:
Whiteside, J.O., Garnsey, S.M., and Timmer, L.W. (eds).
Compendium of Citrus Diseases. APS Press. 80 p.
Garnsey, S.M. and Muller, G.W. 1988. Efficiency of mechanical
transmission of citrus tristeza virus. Pages 46-54, in:
Timmer, L.W., Garnsey, S.M., and Navarro. L. (ed). Proc.
10th Conf. Intern. Organ. Citrus Virol. Riverside,
California.
Garnsey, S.M. and Whidden, R. 1970. A rapid technique for
making leaf tissue grafts to transmit citrus viruses.
Plant Dis. Rep. 54: 907-908.
Garnsey, S.M., Bar-Joseph, M., and Lee, R.F. 1981a.
Applications of serological indexing to develop control
strategies for citrus tristeza virus. Proc. Intern. Soc.
Citriculture 1:448-452.
Garnsey, S.M., Christie, R.G., Derrick, K.S., and Bar-Joseph,
M. 1980a. Detection of citrus tristeza virus. II. Light
and electron microscopy of inclusions and viral
particles. Pages 9-16, in: Calavan, E.C., Garnsey, S.M.,
and Timmer, L.W. (eds). Proc. 8th Conf. Intern. Organ.
Citrus Virol. Riverside, California.
Garnsey, S.M., Gonsalves, D., and Purcifull, D.E. 1977.
Mechanical transmission of citrus tristeza virus.
Phytopathology 67:965-968.
Garnsey, S.M., Gonsalves, D., and Purcifull, D.E. 1979. Rapid
diagnosis of citrus tristeza virus infections by sodium
dodecyl sulphate-immunodiffusion procedures.
Phytopathology 69:88-95.
Garnsey. S.M., Gumpf, D.J., Roistacher, C.N. Civerolo, E.L.,
Lee, R.F., and Yokomi, R.K. 1987. Toward a standardized
evaluation of the biological properties of citrus
tristeza virus. Phytophylactica 19:151-157.
Garnsey, S.M., Lee, R.F., and Brlansky, R.H. 1981b.
Preparation and stability of infectious citrus tristeza
virus (CTV). Phytopathology 71:218 (Abstr.).
Garnsey, S.M., Lee, R.F., Youtsey, C.O., Brlansky, R.H., and
Burnett, H.C.. 1980b. A survey for citrus tristeza virus
in registered budwood sources commercially propagated on
sour orange rootstock in Florida. Proc. Fla. State Hort.
Soc. 93:7-9.

98
Gonsalves, D. and Garnsey, S.M. 1989. Cross-protection
techniques for control of plant virus diseases in the
tropics. Plant Disease 73:592-597.
Gonsalves, D., D.E. Purcifull, and Garnsey, S.M. 1978.
Purification and serology of citrus tristeza virus.
Phytopathology 68:553-559.
Grant, T.J., Moreira, S., and Salybe, A.A. 1961. Citrus
variety reaction to tristeza virus in Brazil when used
in various rootstock and scion combinations. Plant Dis.
Rep. 45:416-421.
Guerri, J., Moreno, P., and Lee, R.F. 1990. Identification of
citrus tristeza virus strains by peptide maps of virion
coat protein. Phytopathology 80:692-698.
Gumpf, D.J., Zheng, G.-Y., Moreno, P., and Diaz, J.M. 1987.
Production and evaluation of specific monoclonal
antibodies to citrus tristeza virus strains.
Phytophylactica 19:159-161.
Hamilton, R.I. 1985. Using plant viruses for disease control.
HortScience 20:848-852.
Hawkes, R. Niday, E. and Gordon, J. 1982. A dot-
immunobinding assay for monoclonal and other antibodies.
Anal. Bioch. 119:142-147.
Hibi, T. and Saito, Y. 1985. A dot-immunobinding assay for the
detection of tobacco mosaic virus in infected tissues.
J. Gen. Virol. 66:1191-1994.
Ieki, H. 1989. The use of cross-protection with mild strains
of citrus tristeza virus (CTV) to control stem pitting
disease of citrus in Japan. Pages 8-14. FFTC Extension
Bulletin No. 284.
Ieki, H. and Yamada, S. 1980. Inactivation of citrus tristeza
virus (CTV) with heat treatment: Heat tolerance and
inactivation of CTV on rootstock-scion combinations.
Pages 220-224, in: Calavan, E.C., Garnsey, and Timmer,
L.W. (eds). Proc. 8th Conf. Intern. Organ. Citrus Virol.
Riverside, California.
Irey, M.S., Permar, T.A., and Garnsey, S.M. 1988.
Identification of severe isolates of citrus tristeza
virus in young field plantings by enzyme-linked
immunosorbent assay. Proc. Fla. State Hort. Soc. 101:73-
76.

99
Koizumi, M. 1986. Citrus tristeza virus: impact and control
by preinoculation in Japan. Pages 148-156, in FFTC Book
Series No. 33.
Koizumi, M. and Kuhara, S. 1984. Protection of preinoculated
citrus trees against tristeza virus in relation to the
virus concentration detected by ELISA. Pages 41-48, in:
Garnsey, S.M., Timmer, L.W., and Dodds, J.A. (eds). Proc.
9th Conf. Intern. Organ. Citrus Virol. Riverside,
California.
Lee, R.F. 1984. Use of double stranded RNAs to diagnose citrus
tristeza virus strains. Proc. Fla. State Hort. Soc.97:53-
56.
Lee, R.F., Brlansky, R.H., and Derrick, K.S. 1988a. Recent
progress on studies of citrus blight: Where do we go from
here. Citrus Industry 69 (2):24,26,29,32,34.
Lee, R.F., Brlansky, R.H., Garnsey, S.M., and Yokomi, R.K.
1987a. Traits of citrus tristeza virus important for mild
strain cross protection of citrus: The Florida approach.
Phytophylactica 19:215-218.
Lee, R.F., Calvert, L.A., Nagel, J., and Hubbard, J.M. 1988b.
Citrus tristeza virus : Characterization of coat
proteins. Phytopathology 78:1221-122.
Lee, R.F., Garnsey, S.M., Brlansky, R.H., and Goheen, A.C.
1987b. A purification procedure for enhancement of citrus
tristeza virus yields and its application to other
phloem-limited viruses. Phytopathology 77:543-549.
Lee, R.F., Garnsey, S.M., Marais, L.J., Moll, J.N., and
Youtsey, C.O. 1988c. Distribution of citrus tristeza
virus in grapefruit and sweet orange in Florida and South
Africa. Pages 33-38, in: Timmer, L.W., Garnsey, S.M. and
Navarro, L. (eds). Proc..10th Conf. Intern. Organ. Citrus
Virol. Riverside, California.
Lister, R.M. and Bar-Joseph, M. 1981. Closteroviruses. Pages
809-844,in: Kurstak, E. (ed). Handbook of Plant Virus
Infections and Comparative Diagnosis. Elsevier/North-
Holland Biomedical Press. Canada.
Marco, G.M. and Gumpf, D.J. 1990. A simple technique for the
production of highly specific polyclonal antisera for
citrus tristeza virus. In: Proc. 11th Conf. Intern.
Organ. Citrus Virol. Riverside, California (in press).

100
McClean, A.P.D. 1957. Tristeza virus of citrus: Evidence for
absence of seed transmission. Plant Disease Reporter 41:
821.
McClean, A.P.D. 1974. The tristeza virus complex. Pages 59-
66, in: Weathers, L.G. and Cohen, M. (eds) Proc. 6th
Conf. Intern. Organ. Citrus Virol. Riverside, California.
Miyakawa, T. 1987. Protection against citrus tristeza seedling
yellows infection in citrus by pre-inoculation with stem
pitting isolates. Phytophylactica 19:193-196.
Miller, T.J. and Stone, H.O. 1978. The rapid isolation of
ribonuclease-free immunoglobulin G by Protein A-Sepharose
affinity chromatography. J. Immunol. Meth. 4:111-125.
Muller, G.W. 1980. Use of mild strains of citrus tristeza
virus (CTV) to re-establish commercial production of
''Pera sweet orange in Sao Paulo, Brazil. Proc. Fla.
State Hort. Soc. 93:62-64.
Muller, G. W. and Costa, A.S. 1977. Tristeza control in Brazil
by preimmunization with mild strains. Proc. Intern. Soc.
Citriculture 3:868-872.
Muller, G.W. and Costa, A.S. 1987. Search for outstanding
plants in tristeza infected orchards: The best approach
to control the disease by preimmunization. Phytohylactica
19:197-198.
Muller, G.W. and Garnsey, S.M,. 1984. Susceptibility of citrus
varieties, species, citrus relatives, and non-rutaceous
plants to slash-cut mechanical inoculation with citrus
tristeza virus (CTV). Pages 33-40, in: Garnsey, S.M.,
Timmer, L.W., and Dodds, A.J. (eds) Proc. 9th Conf.
Intern. Organ. Citrus Virol. Riverside, California.
Muller, G.W., Rezende, J.A.M., and Costa, A.S. 1982.
Preimmunization: An approach to control plant virus
diseases. Proc. 1st Conf. Impact Viral Dis. Rio de
Janeiro, Brazil.
Norman, G., Price, W.C., Grant, T.J., and Burnett, H. 1961.
Ten years of tristeza in Florida. Proc. Fla.State Hort.
Soc. 74:107-111.
Permar, T.A. and Garnsey, S.M. 1990. Comparison of biological
indexing and immunological assays for identifying severe
Florida isolates of citrus tristeza virus. In: Proc.
11th Conf. Intern. Organ. Citrus Virol. Riverside,
California (in press).

101
Permar, T.A., Garnsey, S.M., Gumpf, D.J., and Lee, R.F. 1990.
A monoclonal antibody that discriminates strains of
citrus tristeza virus. Phytopathology 80:224-228.
Powell, C.A. 1987. Detection of three plant viruses by dot-
immunobinding assay. Phytopathology 77:306-309.
Purcifull, D.E. and Batchelor, D.L. 1977. Immunodiffusion
tests with sodium dodecyl sulfate (SDS)-treated plant
virus and plant inclusions. Technical Bulletin 788.
Florida Agricultural Experiment Station. 39 p.
Rocha-Pea, M.A. and Lee, R.F. 1991. Serological techniques
for detection of citrus tristeza virus. J. Virol. Meth.
(in press).
Rocha-Pea, M.A., Lee, R.F., Permar, T.A., Yokomi, R.K., and
Garnsey, S.M. 1990. Use of enzyme-linked immunosorbent
and dot-immunobinding assays to evaluate two cross
protection experiments after challenge with a severe
citrus tristeza virus isolate. In: Proc. 11th Conf.
Intern. Organ. Citrus Virol. Riverside, California (in
press).
Roistacher, C.N. 1981. A blueprint for disaster: Part II.
Changes in the transmissibility of seedling yellows.
Calif. Citrograph 67:29-32.
Roistacher, C.N., Blue, R.L., Nauer, E.M., and Calavan, E.C.
1974. Suppression of tristeza virus symptoms in Mexican
lime seedlings grown at warm temperatures. Plant Dis.
Rep. 58:757-760.
Roistacher, C.N., Dodds, J.A., and Bash, J.A. 1987. Means of
obtaining and testing protective strains of seedling
yellows and stem pitting tristeza virus: A preliminary
report. Phytophylactica 19:199-203.
Roistacher, C.N., Dodds, J.A., and Bash, J.A. 1988. Cross
protection against citrus tristeza seedling yellows and
stem pitting viruses by protective isolates developed in
greenhouse plants. Pages 91-100, in: Timmer, L.W.,
Garnsey, S.M., and Navarro, L. (eds). Proc. 10th Conf.
Intern. Organ. Citrus Virol. Riverside, California.
Schwartz, R.E. 1968. Transmission of the tristeza virus by a
leaf union method. S. African J. Agr. Sci. 11:617-622.
Spinola, S.M. and Cannon, J.G. 1985. Different blocking agents
cause variation in the immunologic detection of proteins
transferred to nitrocellulose membranes. J. Virol. Meth.
81: 161-165.

102
Still, P.E., Hunter, T.J., Rocha-Pea, M.A., Lee, R.F., and
Niblett, C.L. 1990. Western blotting as a rapid method
for the immunodetection and classification of citrus
tristeza virus isolates. Phytopathology 80: (Abstr.) (in
press).
Thornton, I.R., Emmett, R.W. and Stubbs, L.L. 1980. A further
report on the grapefruit tristeza preimmunization trial
at Mildura, Victoria. Pages 51-53, in: Calavan, E.C.,
Garnsey, S.M., and Timmer, L.W. (eds). Proc. 8th Conf.
Intern. Organ. Citrus Virol. Riverside, California.
Tolba, M.A., Ragab, M.M., and Nour-Eldin, F. 1976. Studies on
citrus tristeza virus disease. II. Distribution and
movement of the causal virus in citrus plants. Pages 63-
67, in: Calavan, E.C. (ed). Proc. 7th Conf. Intern.
Organ. Citrus Virol. Riverside, California.
Tsuchizaki, T., Sasaki, A., and Saito, Y. 1978. Purification
of citrus tristeza virus from diseased citrus fruits and
the detection of the virus in citrus tissues by
fluorescent antibody techniques. Phytopathology 68:139-
142.
Van Vuuren, S.P. and Moll, J.N. 1987. Glasshouse evaluation
of citrus tristeza virus isolates. Phytophylactica
19:219-221.
Vela, C., Cambra, M., Corts, E., Moreno, P., Miguet, J.G.,
Prez de San Romn, C., and Sanz, A. 1986. Production and
characterization of monoclonal antibodies specific for
citrus tristeza virus and their use in diagnosis. J. Gen.
Virol. 67:91-96.
Vela, C., Cambra, M., Sanz, A., and Moreno, P. 1988. Use of
specific monoclonal antibodies for diagnosis of citrus
tristeza virus. Pages 55-61, in: Timmer, L.W., Garnsey,
S.M., and Navarro, L. (eds). Proc. 10th Conf. Intern.
Organ. Citrus Virol. Riverside, California.
Wallace, J.W. 1951. Recent developments in the studies of
quick decline and related diseases. Phytopathology
41:785-793.
Wallace, J.M. 1978. Virus and virus-like diseases. Pages 67-
184, in: Reuter, W., Calavan, E.C., and Carman, E.G.
(eds). The Citrus Industry. Vol IV. Crop Protection.
University of California. 362 p.
Wallace, J.M. and Drake. 1976. Progress report of studies in
California on preimmunization against tristeza in budded

103
citrus. Pages 58-62, in: Calavan, E.C. (ed). Proc. 7th
Conf. Intern. Organ. Citrus Virol. Riverside, California.
Yamaguchi, A. and Patpong, P. 1980. Comparison of time
requirement for graft transmission of citrus tristeza
virus with other fruit tree viruses. Pages 25-27, in:
Calavan, E.C., Garnsey, S.M., and Timmer, L.W. (eds).
Proc. 8th Conf. Intern. Organ. Citrus Virol. Riverside,
California.
Yokomi, R.K. and Garnsey, S.M. 1987. Transmission of citrus
tristeza virus by Aphis qossvpii and Aphis citricola in
Florida. Phytophylactica 19:169-172
Yokomi, R.K., Garnsey, S.M., Lee, R.F., and Cohen, M. 1987.
Use of insect vectors to screen for protecting effects
of mild citrus tristeza virus isolates in Florida.
Phytophylactica 19:183-185.
Yokomi, R.K., Garnsey, S.M., Permar, T.A., Lee, R.F., and
Youtsey, C.O. 1990. Natural spread of severe citrus
tristeza virus in citrus preinoculated with mild
isolates. In: Proc. 11th Conf. Intern. Organ. Citrus
Virol. Riverside, California (in press).

BIOGRAPHICAL SKETCH
Mario Alberto Rocha-Pea was borne in Monterrey, Nuevo
Len, Mxico, on July 30, 1950. He received the degree of
Bachelor of Science in microbiology in 1977 from the
Universidad Autnoma de Nuevo Len, and in 1979 received his
Master of Science degree in plant pathology at the Colegio de
Postgraduados, Chapingo, Mxico. From 1979 to 1983 he served
as a professor in the Department of Plant Pathology at the
Colegio Superior de Agricultura Tropical, Tabasco. In 1984
he accepted a position as a research plant pathologist at the
Instituto Nacional de Investigaciones Forestales y
Agropecuarias (INIFAP), in his home state of Nuevo Len, where
he was working until December 1986, before he came to United
States to pursue the degree of Doctor of Philosophy in plant
pathology at the University of Florida.
104

I certify that I have read this study and that in my opinion,
it conforms to acceptable standards of scholarly presentation and
is fully adequate, in scope and quality, as a dissertation for the
degree of Doctor of Philosophy.
Lee, Chairman
Professor of Plant
Pathology
I certify that I have read this study and that in my opinion,
it conforms to acceptable standards of scholarly presentation and
is fully adequate, in scope and quality, as a dissertation for the
degree of Doctor of Philosophy.
C.L. Niblett, Cochairman
Professor of Plant
Pathology
I certify that I have read this study and that in my opinion,
it conforms to acceptable standards of scholarly presentation and
is fully adequate, in scope and quality, as a dissertation for the
degree of Doctor of Philosophy.
.£. fi
D.E. Purcifull
Professor of PI
Pathology
I certify that I have read this study and that in my opinion,
it conforms to acceptable standards of scholarly presentation and
is fully adequate, in scope and quality, as gf dissertation for the
degree of Doctor of Philosophy.
' S .VM. Garhsey
Professor
Pathology
I certify that I have read this study and that in my opinion,
it conforms to acceptable standards of scholarly presentation and
is fully adequate, in scope and quality, as a dissertation for the
degree of Doctor of Philosophy.
^5- 1C.
R.K. Yokomi
Assistant Professor of
Entomology and Nematology

I certify that I have read this study and that in my opinion,
it conforms to acceptable standards of scholarly presentation and
is fully adequate, in scope and quality, as a dissertation for the
deqree of Doctor of Philosophy.
December 1990
Deant
Agriculture
ollege of^
Dean, Graduate School



8
protection against either seedling yellows or stem pitting CTV
isolates from a total of 116 evaluated when challenged with
the original isolates from which they were derived. This
approach required extensive greenhouse work to follow the
segregation of every single isolate, and some attenuated
isolates showed a tendency to revert back to their severe
forms.
Recently, the greenhouse evaluation of CTV-stem pitting
isolates as protecting agents against the CTV-ID on sweet/sour
orange combinations was reported (Miyakawa, 1987). However,
CTV-stem pitting isolates as protecting agents offer limited
possibilities of commercial use in Florida, particularly
because stem pitting isolates still are not present and
susceptible species or cultivars such as grapefruit are grown
extensively.
Citrus tristeza virus causing decline on trees on sour
orange rootstock is endemic in Florida. The effective use of
CTV-tolerant rootstocks is diminished by their susceptibility
to citrus blight. In addition, the increased popularity of
sour orange despite tristeza, along with the natural
prevalence of mild CTV isolates, provides the opportunity to
evaluate cross protection as an alternative control strategy
for CTV-ID. However, there are several considerations: i)
The threat of recurrent freezes makes it difficult to reliably
evaluate the cross protection potential of mild isolates under
field conditions, especially in the Ridge area; ii) There is


90
of recurrent freezes that can hamper the reliable evaluation
of cross-protection experiments under field conditions.
Graft Transmission of Citrus Tristeza Virus
The efficiency of graft transmission of CTV by leaf or
bark pieces was conditioned primarily by the donor/receptor
host combination and secondly by the virus isolate involved.
For example, C. excelsa showed a low 72.4% and 60.7% rate of
transmission to Madam Vinous and grapefruit, respectively;
whereas, an 86.7% rate of transmission was obtained to Mexican
lime plants. In regard to Mexican lime as donor hosts, there
was a rate of transmission of 89.3% and 93.1.7% to grapefruit
and Madam Vinous, respectively and a 76.9% rate to Mexican
lime. Rate of transmission from Madam Vinous sweet orange was
between 84.6 and 100% in all receptors tested. Of the three
donor hosts tested, Madam Vinous sweet orange was the most
efficient donor host (90.6%), followed by Mexican lime
(85.2%). C. excelsa was a poor donor host with an overall
rate of transmission of 72.5%, being relatively efficient only
when inoculated to Mexican lime. Differences were found
in the rate of transmission of the three different CTV
isolates with each donor host tested. However, the overall
rate of transmission was in the range of 80% for all isolates.
Apparently the efficiency of the graft transmission of the
virus depended more on the donor host involved than the virus
isolate.


99
Koizumi, M. 1986. Citrus tristeza virus: impact and control
by preinoculation in Japan. Pages 148-156, in FFTC Book
Series No. 33.
Koizumi, M. and Kuhara, S. 1984. Protection of preinoculated
citrus trees against tristeza virus in relation to the
virus concentration detected by ELISA. Pages 41-48, in:
Garnsey, S.M., Timmer, L.W., and Dodds, J.A. (eds). Proc.
9th Conf. Intern. Organ. Citrus Virol. Riverside,
California.
Lee, R.F. 1984. Use of double stranded RNAs to diagnose citrus
tristeza virus strains. Proc. Fla. State Hort. Soc.97:53-
56.
Lee, R.F., Brlansky, R.H., and Derrick, K.S. 1988a. Recent
progress on studies of citrus blight: Where do we go from
here. Citrus Industry 69 (2):24,26,29,32,34.
Lee, R.F., Brlansky, R.H., Garnsey, S.M., and Yokomi, R.K.
1987a. Traits of citrus tristeza virus important for mild
strain cross protection of citrus: The Florida approach.
Phytophylactica 19:215-218.
Lee, R.F., Calvert, L.A., Nagel, J., and Hubbard, J.M. 1988b.
Citrus tristeza virus : Characterization of coat
proteins. Phytopathology 78:1221-122.
Lee, R.F., Garnsey, S.M., Brlansky, R.H., and Goheen, A.C.
1987b. A purification procedure for enhancement of citrus
tristeza virus yields and its application to other
phloem-limited viruses. Phytopathology 77:543-549.
Lee, R.F., Garnsey, S.M., Marais, L.J., Moll, J.N., and
Youtsey, C.O. 1988c. Distribution of citrus tristeza
virus in grapefruit and sweet orange in Florida and South
Africa. Pages 33-38, in: Timmer, L.W., Garnsey, S.M. and
Navarro, L. (eds). Proc..10th Conf. Intern. Organ. Citrus
Virol. Riverside, California.
Lister, R.M. and Bar-Joseph, M. 1981. Closteroviruses. Pages
809-844,in: Kurstak, E. (ed). Handbook of Plant Virus
Infections and Comparative Diagnosis. Elsevier/North-
Holland Biomedical Press. Canada.
Marco, G.M. and Gumpf, D.J. 1990. A simple technique for the
production of highly specific polyclonal antisera for
citrus tristeza virus. In: Proc. 11th Conf. Intern.
Organ. Citrus Virol. Riverside, California (in press).


9
a need for a method to more rapidly select and evaluate
potential cross protecting mild isolates which is adaptable
for screening large numbers of isolates; and iii) There is
a need to differentiate mild and severe CTV isolates in a
mixed infection in the same plant to aid in the evaluation and
to better understand the mechanism by which cross protection
functions.
The objectives of this research were to evaluate
naturally occurring Florida mild CTV isolates for cross
protecting ability and to develop a methodology to detect the
presence of protecting and challenge CTV isolates in mixed
infections. The effect of temperature on the cross protecting
ability of CTV mild isolates also was studied.
Materials and Methods
Virus isolates and donor hosts. Five naturally occurring
Florida CTV isolates collected from field grown sweet orange
or grapefruit trees grafted on sour orange were used in these
experiments after transmission by Aphis qossypii Glover. The
Tila, T26, T30 and T55a isolates produce very mild symptoms
and little or no stunting on Mexican lime seedlings {Citrus
aurantifolia (Christm.) Swingle}, no seedling yellows on
Eureka lemon {C. limn (L.) Burm.} or sour orange (C.
aurantium L.) and are symptomless in sweet orange {C. sinensis
(L.) Osb.} and sweet/sour orange combinations (Garnsey et al.
1987; Lee, 1984; Yokomi and Garnsey, 1987; Yokomi et al.
1987) The T66a challenge isolate causes strong vein


64
reaction on Mexican lime than isolates listed above. The
isolates T3, T36, T62a, T65a, T66a and T67a cause strong vein
clearing, stunting and stem pitting in Mexican lime seedlings,
varying degrees of decline and/or growth reduction in
sweet/sour orange combinations, varying degrees of seedling
yellow reaction and/or severe stunting in sour orange or
grapefruit seedlings, and varying degrees of growth reduction
in Madam Vinous sweet orange seedlings.
The relative sensitivity limits were compared for
detection of CTV by DIBA, DAS-ELISA, and DAS-indirect ELISA.
An extract {1:10 (w/v) in TBS-Tween} was made of greenhouse
grown C. excelsa plants infected with the CTV isolate T36.
A series of two fold dilutions was made with a similar extract
of healthy C. excelsa. Negative controls included TBS-Tween,
and 1:10 extracts of healthy C. excelsa. Madam Vinous sweet
orange, Duncan grapefruit and Mexican lime plants, all in TBS-
Tween.
Evaluation of different buffers for sample extraction.-
An additional experiment was carried out to determine the
effects of different extraction buffers on the sensitivity of
DIBA, DAS-ELISA, and DAS-indirect ELISA. Three different
extraction solutions, Tris buffered saline, phosphate buffered
saline, and carbonate buffer (all as described above), with
and without 0.05% Tween 20, were used. Three CTV isolates
T26, T62a, and T66a propagated in Madam Vinous sweet orange
plants were tested in DIBA, DAS-ELISA, and DAS-indirect ELISA


87
compared with sensitivities of double antibody sandwich (DAS)
ELISA and DAS-indirect ELISA with 12 different CTV isolates.
The conclusions of this research are summarized as follows:
Cross-Protection of Citrus Tristeza Virus
At warm temperatures plants pre-inoculated with mild
isolates showed relatively low optical density (OD405) values
in the range of 0.091-0.145 in DAS-ELISA with polyclonal
antisera; whereas, at cool temperatures the values were
commonly in the range of 0.300-0.400. At warm temperatures,
the MCA-13 monoclonal antibody in plants pre-inoculated with
mild isolates but unchallenged, gave OD405 values in the range
of 0.030-0.045 which were as low as the uninoculated control
plants. However, at cool temperatures, the OD405 values
obtained with mild isolates were slightly higher than those
obtained at warm temperatures, indicating that the MCA-13
monoclonal antibody may react to some extent with mild
isolates when they are above a certain titer in the plants.
Nevertheless, the OD405 values were always lower than 0.100,
which was considered a negative reaction.
When plants pre-inoculated with mild isolates and further
challenged with the severe isolate were assayed with the MCA-
13 monoclonal antibody, they gave typically lower OD405 values
for the T66a isolate than the unprotected challenged control
plants which did not have mild isolates. In this regard, the
T26 and T3 0 isolates gave the more uniform lower OD405 values
for the T66a isolate in all evaluated treatments. This would


68
A
B
C
D
i


2 t
;
3
4
Fig. 4.2 Reaction of polyclonal and monoclonal antibodies in
dot-immunobinding assay with citrus tristeza virus (CTV)
isolates. A. Polyclonal antibodies 1053 (1.0 /g/ml) not
cross-absorbed with buffer extract of healthy plant. B.
Polyclonal antibodies 1053 (1.0 /g/ml) incubated with 1:200
(v/v) buffer extract from healthy plants for 2 hr at 37C
prior to use. C and D, monoclonal 3DF1 (1.0 /ig/ml) and MCA-
13 (1:5,000 dilution) antibodies not cross-absorbed with
healthy extract. Number 1, CTV T66a; 2, CTV T26; 3, buffer
extract from healthy sweet orange plants; 4, TBS-Tween.
Reaction conditions are described in materials and methods.


I certify that I have read this study and that in my opinion,
it conforms to acceptable standards of scholarly presentation and
is fully adequate, in scope and quality, as a dissertation for the
degree of Doctor of Philosophy.
Lee, Chairman
Professor of Plant
Pathology
I certify that I have read this study and that in my opinion,
it conforms to acceptable standards of scholarly presentation and
is fully adequate, in scope and quality, as a dissertation for the
degree of Doctor of Philosophy.
C.L. Niblett, Cochairman
Professor of Plant
Pathology
I certify that I have read this study and that in my opinion,
it conforms to acceptable standards of scholarly presentation and
is fully adequate, in scope and quality, as a dissertation for the
degree of Doctor of Philosophy.
.£. fi
D.E. Purcifull
Professor of PI
Pathology
I certify that I have read this study and that in my opinion,
it conforms to acceptable standards of scholarly presentation and
is fully adequate, in scope and quality, as gf dissertation for the
degree of Doctor of Philosophy.
' S .VM. Garhsey
Professor
Pathology
I certify that I have read this study and that in my opinion,
it conforms to acceptable standards of scholarly presentation and
is fully adequate, in scope and quality, as a dissertation for the
degree of Doctor of Philosophy.
^5- 1C.
R.K. Yokomi
Assistant Professor of
Entomology and Nematology


29
host for purification purposes of CTV (Lee, et al. 1987b,
1988b) However, it had never been used for leaf graft
transmission experiments. At least three inoculations with
C. excelsa tissue infected with mild isolates were made
unsuccessfully, even though the inoculated tissue survived for
at least four weeks post-inoculation and was left in place
for several months. It was necessary to switch to Madam
Vinous sweet orange and/or Mexican lime plants infected with
the mild isolates as inoculum sources before a minimum of 3
or 4 plants per treatment were positively infected with mild
isolates as determined by DAS-ELISA. Furthermore, some
treatments, mostly Valencia/macrophylla, were not evaluated
because of the lack of replications because not enough plants
were available. This originated the need to design another
separate experiment, described in Chapter 3, to determine the
effectiveness of different citrus species as donor hosts for
graft transmission of the virus.
At warm temperatures, plants inoculated with mild
isolates but unchallenged with T66a had relatively low OD405
values in the range of 0.095-0.145 (Tables 2.1 and 2.2). At
cool temperatures, with few exceptions, were commonly higher
in the range of 0.300-0.400 (Tables 2.3 and 2.4). At warm
temperatures there were some treatments that gave OD405 values
lower than 0.100, which may be interpreted as negative
reactions. However, those values were the averages of the
OD405 readings of every treatment. Thus a single plant with a


Table 2.5 Effect of citrus tristeza virus (CTV) mild isolates on the development of
the CTV decline syndrome in plants unchallenged and challenged by the T66a severe
isolate: I. Warm temperature, Valencia/sour orange.
Unchallenged
1/
Challenged
1/2/
CTV
No. of
Decline
index'
No. of
Decline
No. of dead
isolate
plants
plants
index
plants 10 months
after challenge
Tila
4
0.0
8
6.3
3/8
T26
4
1.5
5
2.6
0/5
T30
5
1.0
4
4.2
1/4
T55a
5
0.0
7
5.5
2/7
Healthy
or control
5
1.4
6
4.0
1/6
plants uninoculated
with mild isolate
/ One year
old plants were graft inoculated
with leaf
pieces under bark flaps on the
stem from donor
plants infected
with the indicated
mild CTV isolates.
/ After verifying
virus infection
with mild
isolates
by DAS-ELISA with polyclonal
to
u
antibodies, the challenged plants were graft inoculated similarly with a T66a severe
isolate.
Decline index is the average per treatment of the cumulative score given by visual
readings on three criteria: stem diameter, growth reduction, and foliage decline
symptoms, where 0 = healthy vigorous plant to 3 = severely diseased. Minimum index
= 0, maximum index =9. A high decline index sometimes was accompanied by plant
death.
2/


58
chickens (Bar-Joseph and Malkinson, 1980; Marco and Gumpf,
1990). Likewise, several monoclonal antibodies (MCAs) have
been developed in mice (Gumpf et al. 1987; Permar et al. 1990;
Vela et al. 1986, 1988). The 3DF1 MCA has been reported to
react with a broad spectrum of CTV isolates of different
geographical origins and is available commercially (Vela et
al. 1986, 1988). The MCA-13 MCA (hereafter MCA-13) has been
reported to react specifically with CTV isolates that have
severe biological activities, especially isolates causing
decline on plants grafted on sour orange rootstock (Garnsey
and Permar, 1990; Permar et al. 1990). It has been used in
diverse studies for strain discrimination purposes (Irey et
al. 1988; Garnsey and Permar 1990; Rocha-Pea et al. 1990;
Yokomi et al. 1990).
The ELISA test, particularly the double antibody sandwich
system (DAS-ELISA) (Bar-Joseph et al. 1979b; 1980) has been
the most widely used of all serological methods developed for
CTV detection (Bar-Joseph et al. 1989; Garnsey, et al. 1981a;
Rocha-Pea and Lee, 1991) In DAS-ELISA the virus in the test
sample is trapped and immobilized selectively by specific
antibodies adsorbed on polystyrene microtiter plates. Enzyme-
conjugated antibodies are then reacted with the trapped virus
and detected colorimetrically after adding a suitable
substrate (Clark and Adams, 1977; Garnsey and Cambra, 1990).
DAS-ELISA is relatively easy to perform and is highly
sensitive, but it does require some special equipment, it is


44
tissue in the whole experiment was 83 and 66 percent for leaf
and bark pieces, respectively. Overall at least one of the
four grafts survived in 92 and 90 percent of the receptor
plants inoculated with leaf or bark pieces, respectively. In
calculating the percent of virus transmission for each
donor/receptor host/virus isolate combination, only those
plants with at least one (of four) surviving inoculum piece
were taken into account. Thus, overall there was a greater
efficiency of transmission with leaf pieces (89.2%) than with
bark pieces (75.6%) for the whole experiment (Table 3.1).
There were three plants of 270 in the entire experiment, one
Mexican lime and two grapefruit that became infected even
though no successful graft was scored 21 days post
inoculation.
The overall rate of transmission of CTV by graft
inoculation for each donor/receptor host combination is shown
in Table 3.2. With C. excelsa as donor host there was 72.4%,
86.9%, and 60.7% transmission to Madam Vinous, Mexican lime
and grapefruit, respectively. With Mexican lime as donor
host, there was 93.1%, 76.9% and 89.3% transmission to Madam
Vinous, Mexican lime and grapefruit, respectively. With Madam
Vinous as donor host there was 86.7%, 100%, and 84.6%
transmission to Madam Vinous, Mexican lime and grapefruit
respectively. The overall average of transmission was 72.5%
from C. excelsa. 85.2% from Mexican lime, and 90.6% from Madam
Vinous (Table 3.2). Statistical analysis showed significant



PAGE 1

CITRUS TRISTEZA VIRUS: CROSS-PROTECTION, GRAFT TRANSMISSION, AND SEROLOGY BY MARIO ALBERTO ROCHA-PENA A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 1990

PAGE 2

To my beloved wife Alicia, and to my adored son Eric, whom are the world of my life. To my parents Herminio and Dolores, and my brothers and sisters.

PAGE 3

ACKNOWLEDGEMENTS I want to express my sincere gratitude and acknowledgment to the following persons, institutions, and organizations, who supported my graduate studies at the University of Florida. The financial support from Consejo Nacional de Ciencia y Tecnologia (CONACyT) , as well as the support and leave of absence from Instituto Nacional de Investigaciones Forestales y Agropecuarias (INIFAP) , both institutions from Mexico, is greatly appreciated. To my co-major professors, Drs. R.F. Lee and C.L. Niblett, for their constant encouragement, guidance and support throughout the course of the thesis work, and for their personal interest in my academic preparation. To them and the rest of the supervisory committee, Drs. S.M. Garnsey, D.E. Purcifull, and R.K. Yokomi, for valuable suggestions in revision of the manuscript. The financial support from the Florida High Technology and Industrial Council, and the Florida Citrus Production Managers' Association, to carry out some parts of the thesis work also is acknowledged. I thank Drs. S.M. Garnsey and T.A. Permar, USDA Orlando, and Drs. P. Moreno and M. Cambra, IVIC Valencia (Spain), for supplying the MCA-13 and 3DF1 monoclonal antibodies, respectively. ii

PAGE 4

The technical assistance and friendship of N. Berger, S. Marquardt, T. Nguyen, S. Jackson, and J. Zellers, as well as the help of T. Zito in the photographic work, and M. Ahnger with the statistical analysis, all at the Citrus Research and Education Center, at Lake Alfred, is greatly appreciated. The warm friendship I found in my fellow students, lab technicians, administrative staff, and most faculty members at the Plant Pathology Department, Gainesville, and at the Citrus Research and Education Center, Lake Alfred, was indeed encouraging and will be unforgettable. Finally, I want to thank my wife Alicia for her constant encouragement and patience to endure the hardship of my pursuit of this graduate degree. iii

PAGE 5

TABLE OF CONTENTS Page ACKNOWLEDGMENTS ii LIST OF FIGURES vi LIST OF TABLES ix ABSTRACT xi CHAPTER 1. INTRODUCTION 1 2. EVALUATION OF THE PROTECTING EFFECTS OF SOME FLORIDA ISOLATES OF CITRUS TRISTEZA VIRUS AGAINST THE DEVELOPMENT OF THE DECLINE SYNDROME 6 Introduction 6 Materials and Methods 9 Virus isolates and donor hosts 9 Inoculation of CTV isolates and receptor hosts. 10 Serological tests 11 Detrimental effects of mild isolates and evaluation of cross-protection 14 Results 14 Antigen titers of mild and severe CTV isolates with polyclonal and MCA-13 monoclonal antibodies 14 Detrimental effect of mild isolates and evaluation of cross-protection 21 Discussion 27 3. EFFECTIVENESS OF DIFFERENT CITRUS SPECIES AS DONOR HOSTS FOR GRAFT TRANSMISSION OF CITRUS TRISTEZA VIRUS 38 Introduction 38 Materials and Methods 40 Virus isolates and donor hosts 40 Grafting procedures and receptor hosts 41 Virus distribution and antigen concentration in host tissues 41 Purification of CTV 42 iv

PAGE 6

Serological tests 42 Results 43 Graft transmission of citrus tristeza virus isolates 43 Virus distribution and antigen concentration in host tissues 47 Discussion 52 4. DEVELOPMENT OF A DOT-IMMUNOBINDING ASSAY FOR CITRUS TRISTEZA VIRUS 57 Introduction 57 Materials and Methods 59 Antisera used 59 Sample preparation 60 Dot-immunobinding assay 61 DAS -ELI SA 62 DAS-indirect ELISA 63 Evaluation 63 Evaluation of different buffers for sample extraction 64 Results 65 Development of the dot-immunobinding assay 65 Evaluation 69 Evaluation of different buffers for sample extraction 77 Discussion 77 5. SUMMARY AND CONCLUSIONS 86 LITERATURE CITED 94 BIOGRAPHICAL SKETCH 104 v

PAGE 7

LIST OF FIGURES Page Figure 2.1 Plot of purified citrus tristeza virus (CTV) against optical density. Bark of healthy Citrus excelsa (0.5 g) tissue was ground in 5.0 ml of phosphate buffered saline, pH 7.6, + 0.05% Tween + 2% polyvinyl pyrrolidone and mixed with purified CTV T26 isolate to give the desired optical density at 260 nm (OD 260 ) . DAS ELI S A was performed as described in materials and methods. An extinction coefficient of 2.0 was assumed (Gonsalves et al. 1978) to estimate the relative virus concentration 22 Figure 3 . 1 Plot of purified citrus tristeza virus (CTV) against optical density. Bark of healthy Citrus excelsa (0.25 g) tissue was ground in 5.0 ml of phosphate buffered saline, pH 7.6, + 0.05% Tween + 2% polyvinyl pyrrolidone and mixed with purified CTV T26 isolate to give the desired optical density at 260 nm (OD 260 ) . DAS-ELI SA was performed as described in materials and methods. An extinction coefficient of 2.0 was assumed (Gonsalves et al. 1978) to estimate the relative virus concentration 51 Figure 4.1 Effect of different blocking solutions on the reaction of polyclonal antibodies no. 1053 in dot-immunobinding assay with citrus tristeza virus (CTV) isolates: A. TBS alone; B. 3% bovine serum albumin (BSA) ; C. 3% gelatin; D. 0.5% non-fat dry milk; E. 5% Triton X-100. Number 1, CTV T66a; 2, CTV T26; 3, buffer extract from healthy sweet orange plants; 4, TBS-Tween. Reaction conditions are described in materials and methods 66 Figure 4.2 Reaction of polyclonal and monoclonal antibodies in dot-immunobinding assay with citrus tristeza virus (CTV) isolates. A. Polyclonal 1053 (1.0 /ug/ml) not crossvi

PAGE 8

absorbed with buffer extract of healthy plant. B. Polyclonal 1053 (1.0 tig/ml) incubated with 1:200 (v/v) buffer extract from healthy plants for 2 hr at 37°C prior to use. C and D, monoclonal 3DF1 (1.0 /xg/ml) and MCA-13 (1:5,000 dilution) antibodies not cross-absorbed with healthy extract. Number 1, CTV T66a; 2, CTV T26; 3, buffer extract from healthy sweet orange plants; 4, TBS-Tween. Reaction conditions are described in materials and methods 68 Figure 4.3 Relative sensitivity level of different polyclonal and monoclonal antibodies specific to citrus tristeza virus (CTV) in dot-immunobinding assay. Rows A, B and C, polyclonal antibodies nos. 1051, 1052, and 1053, respectively. Rows D and E, monoclonal antibodies 3DF1 and MCA-13, respectively. Extract of Citrus excelsa greenhouse grown plants infected with CTV T36 isolate was prepared in TBS-Tween 1:10 (w/v) and two fold diluted with extract of healthy plants. Reaction conditions are described in materials and methods 70 Figure 4.4 Reaction of polyclonal antibodies no. 1053 and 3DF1 and MCA-13 monoclonal antibodies in dot-immunobinding assay with twelve selected citrus tristeza virus (CTV) isolates. Row A: 1 = Tlla; 2 = T26; 3 = T30; 4 = T50a; 5 = T55a; 6 = T3; 7 = T4; 8 = T36. ROW B. 1 = T62a; 2 = T65a; 3 = T66a; 4 = T67a. Also in Row B are extracts of healthy plants: 5 = Citrus excelsa ; 6 = Madam Vinous sweet orange; 7 = Duncan grapefruit; 8 = Mexican lime. ... 74 Figure 4.5 Evaluation of different extraction buffers on the sensitivity of DIBA with citrus tristeza virus (CTV) isolates T26, T62a and T66a. TBS = Tris buffered saline, TBST = TBS containing 0.05% Tween, PBS = phosphate buffered saline, PBST = PBS + 0.05% Tween, Carb = carbonate buffer, CarbT = Carb + 0.05% Tween, HC = healthy control. The IgG of polyclonal antibodies No. 1053 or monoclonal antibody 3DF1 were used as primary antibodies followed by the goat antirabbit and antimouse IgG conjugates, respectively. Samples were 2 /xl of buffer VII

PAGE 9

extracts of greenhouse-grown Madam Vinous sweet orange plants infected with the CTV isolates ground at a 1:10 dilution 78 Figure 4.6 Evaluation of different extraction buffers on the sensitivity of DAS-ELISA and DAS-indirect ELISA with citrus tristeza virus (CTV) isolates T26, T62a and T66a. TBS = Tris buffered saline, TBST = TBS+Tween (0.05%), PBS = phosphate buffered saline, PBST = PBS+Tween (0.05%), Carb = carbonate buffer, CarbT = Carb+Tween (0.05%), HC = healthy control. The IgG of polyclonal antibodies No. 1053 was used to coat the plates for both DAS-ELISA and DAS-indirect ELISA. For DAS-ELISA the No. 1053 IgG conjugate was used as second antibody. For DAS-indirect ELISA the unlabeled 3DF1 monoclonal antibody was the intermediate antibody followed by the goat anti-mouse IgG conjugate. Samples were 200 /il of buffer extracts of greenhouse-grown Madam Vinous sweet orange plants infected with the CTV isolates ground at a 1:10 dilution 79 viii

PAGE 10

LIST OF TABLES Page Table 2.1 Table 2.2 Relative antigen titer of citrus tristeza virus (CTV) mild isolates in plants unchallenged and challenged by the T66a severe isolate: I. Warm temperature, Valencia/sour orange , 16 Relative antigen titer of citrus tristeza virus (CTV) mild isolates in plants unchallenged and challenged by the T66a severe isolate: II. Warm temperature, Valencia/macrophylla , 17 Table 2.3 Table 2.4 Table 2.5 Table 2.6 Relative antigen titer of citrus tristeza virus (CTV) mild isolates in plants unchallenged and challenged by the T66a severe isolate: III. Cool temperature, Valencia/sour orange 19 Relative antigen titer of citrus tristeza virus (CTV) mild isolates in plants unchallenged and challenged by the T66a severe isolate: IV. Cool temperature, Valencia/macrophylla , 20 Effect of citrus tristeza virus (CTV) mild isolates on the development of the CTV decline syndrome in plants unchallenged and challenged by the T66a severe isolate: I. Warm temperature, Valencia/ sour orange 23 Effect of citrus tristeza virus (CTV) mild isolates on the development of the CTV decline syndrome in plants unchallenged and challenged by the T66a severe isolate: II. Warm temperature, Valencia/macrophylla 25 ix

PAGE 11

Table 2.7 Effect of citrus tristeza virus (CTV) mild isolates on the development of the CTV decline syndrome in plants unchallenged and challenged by the T66a severe isolate: III. Cool temperature, Valencia/ sour orange 26 Table 2.8 Effect of citrus tristeza virus (CTV) mild isolates on the development of the CTV decline syndrome in plants unchallenged and challenged by the T66a severe isolate: IV. Cool temperature, Valencia/macropylla 28 Table 3.1 Transmission of citrus tristeza virus by graft inoculation between selected citrus hosts: I. Efficiency of leaf and bark pieces as inoculum 45 Table 3.2 Transmission of citrus tristeza virus by graft inoculation between selected citrus hosts: II. Overall rate of transmission 46 Table 3.3 Transmission of citrus tristeza virus by graft inoculation between selected citrus hosts: III. Effect of virus isolates 48 Table 3 . 4 Relative antigen titer of citrus tristeza virus in different tissues of Citrus excelsa and Madam Vinous sweet orange host plants, as measured by enzymelinked immunosorbent assay 50 Table 4.1 Relative sensitivity of DIBA, DASELI SA and DAS-indirect with polyclonal (1053) and monoclonal (3DF1 and MCA-13) antibodies specific to citrus tristeza virus 71 Table 4.2 Comparison of DIBA, DAS-ELISA and DAS-indirect ELISA for the detection of citrus tristeza virus (CTV) and relative reactivity of polyclonal (1053) and monoclonal (3DF1) (MCA-13) antibodies with CTV isolates 75 x

PAGE 12

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy CITRUS TRISTEZA VIRUS: CROSS-PROTECTION, GRAFT TRANSMISSION, AND SEROLOGY By MARIO ALBERTO ROCHA-PENA DECEMBER 1990 Chairperson: Dr. Richard F. Lee Major Department: Plant Pathology The objectives of this research were i) to evaluate some citrus tristeza virus (CTV) mild isolates under greenhouse conditions for cross protecting ability against the decline syndrome, and ii) to develop methods for detection of severe CTV challenge isolates in mixed infections in cross-protection experiments. Valencia sweet orange plants budded on sour orange rootstock were graft-inoculated by leaf pieces using any of four different mild CTV isolates and subsequently graft-challenged with a severe CTV isolate. Treatments were evaluated at temperature regimens of 21-38°C and 21-33°C. Plants preinoculated with mild isolates when challenged with the severe isolate gave relatively lower ELISA values as compared to the unprotected, challenged control plants. The MCA-13 monoclonal antibody provided a rapid method to detect xi

PAGE 13

the severe isolate in mixed infections. CTV-induced decline (CTV-ID) occurred irregularly within the first 10 months after challenge inoculation at both temperature regimens. The preliminary evaluation of the cross-protecting ability of mild isolates against the CTV-ID in plants on sour orange rootstock can be accomplished under greenhouse conditions in a relatively short time of 18-24 months. Differences were found in the effectiveness of certain tissues and/or hosts for graft-transmission of CTV. Leafpiece grafts transmitted CTV at a 90% rate vs a 75% rate using bark pieces. Madam Vinous sweet orange was the most efficient donor host giving 90% transmission to three receptor hosts, followed by Mexican lime at 85%, and Citrus excelsa at 72%. The dot-immunobinding assay (DIBA) was adapted for CTV diagnosis by using several polyclonal and monoclonal antibodies specific for CTV. The DIBA was as sensitive as DAS ELI S A and DAS-indirect ELISA for CTV detection and provides a reliable alternative for diagnosis of CTV. xii

PAGE 14

CHAPTER 1 INTRODUCTION Citrus tristeza virus (CTV) is distributed in citrus producing areas worldwide and is the most economically important viral disease of citrus (Bar-Joseph et al. 1979a, 1981, 1989) . The virus infects nearly all species, varieties, and intergeneric hybrids of citrus, and some citrus relatives (Bar-Joseph et al. 1979a; Garnsey and Lee, 1988; Muller and Garnsey, 1984) . However, the most destructive damage is the induced decline in scions grafted on sour orange ( Citrus aurantium L. ) rootstock. Some CTV isolates cause stem pitting and loss of plant vigor on some orange and grapefruit scions regardless of the rootstock (Bar-Joseph et al. 1979a, 1981, 1989) . Citrus tristeza virus is a phloem-limited, flexuous closterovirus approximately 2,000 x 11 nm in size, transmitted by aphids in a semi-persistent manner (Bar-Joseph et al. 1979a; Lister and Bar-Joseph, 1981) . A single stranded positive sense RNA of 5.4-6.5 x 10 6 daltons has been isolated from purified virus preparations (Bar-Joseph et al. 1985) . Several coat proteins of about Mr 28,000 (Guerri et al. 1990),

PAGE 15

23,000 and 21,000 (Lee et al. 1988b), respectively, have been associated with CTV virions. The virus is readily transmitted by budding and grafting (Bennett and Costa, 1949; Bar-Joseph and Lee, 1990; Bar-Joseph et al. 1979a) . Mechanical transmission has been accomplished by slash inoculation of partially purified virus preparations into the stem of hosts such as citron ( Citrus medica L.) and Mexican lime {C. aurantifolia (Christm.) Swing.} (Garnsey and Muller, 1988; Garnsey et al. 1977; Muller and Garnsey, 1984). Seed transmission has not been demonstrated (McClean, 1957; Wallace, 1978) . Citrus tristeza virus occurs naturally with a diversity of isolates or strains which may differ greatly in their biological properties, such as symptomatology in different citrus hosts (Garnsey et al. 1987; McClean, 1974), aphid transmissibility ( BarJoseph and Loebenstein, 1973; Bar-Joseph et al. 1977; Roistacher, 1981; Yokomi and Garnsey, 1987), and sensitivity to warm temperatures (Ieki and Yamada, 1980; Roistacher et al. 1974). Several sensitive and relatively rapid methods have been developed to diagnose the presence of CTV in infected plants. These methods include SDS-immunodif fusion procedures (Garnsey et al. 1979; Bar-Joseph et al. 1980), enzyme-linked immunosorbent assay (ELISA) (Bar-Joseph et al. 1979b, 1980) , light and electron microscopy (Brlansky, 1987; Brlansky et al. 1984; Garnsey et al. 1980a), and in situ immunofluorescence

PAGE 16

(Brlansky et al. 1984; Tsuchizaki et al. 1978). Each of these methods has different advantages, disadvantages, and sensitivity levels. Therefore, a certain method may be used for a particular purpose. Some methods, such as ELISA, are dependable and widely used for indexing purposes (Garnsey et al. 1981a). Recently a monoclonal antibody was developed against a decline-inducing isolate from Florida which, in ELISA, reacted specifically with several severe CTV isolates from diverse geographical areas, but not with mild isolates from the same areas (Permar et al. 1990). Control of CTV is difficult. In those few areas of the world where CTV still is not present, quarantine and virusfree certification programs are maintained to prevent the introduction of infected budwood sources (Bar-Joseph et al. 1983, 1989) . Likewise, in those areas with low CTV incidence, large scale surveys and suppression measures are carried out to reduce disease spread to other trees and locations and prolong the use of sour orange as a rootstock (Bar-Joseph et al. 1989) . Once the disease becomes endemic, two situations can result: a) CTV-induced decline develops and kills plants grafted onto sour orange rootstock, whereas tolerant rootstocks do not decline, and/or b) CTV-stem pitting can affect sweet orange and/or grapefruit scions regardless of the rootstock, resulting in a loss of plant vigor and yield. Mild strain cross protection is the only known control measure

PAGE 17

which is effective against stem pitting (Bar-Joseph et al. 1989; Garnsey and Lee, 1988; Lee et al. 1987a). Genetic resistance to CTV is not available in commercially acceptable scions (Bar-Joseph et al. 1989; Garnsey and Lee, 1988). The application of genetically engineered cross protection, currently effective in several other crops (Beachy et al. 1987) , is an attractive possibility for CTV control in the future. CTV has been widespread in Florida for many years and induced decline has occurred in localized areas (Garnsey and Jackson, 1975; Norman et al. 1961). However, until recently, it had not caused major losses because most of the citrus acreage had been propagated on CTV-tolerant rootstocks and because of the prevalence of mild CTV isolates which did not seriously affect trees grafted on sour orange rootstock (Brlansky et al. 1986; Garnsey et al. 1980b; Lee et al. 1987a) . In the last decade the situation in Florida has changed radically. Sour orange continued to be a very popular rootstock because of cold tolerance, high fruit quality of the scion, and its tolerance to citrus blight. The high demand for plants on sour orange, plus discovery of citrus bacterial leaf spot in some nurseries (Brlansky, 1988; Garnsey, personal communication) , caused nurserymen to use budwood from source trees that had not been propagated previously on sour orange. Many such trees apparently were harboring severe CTV isolates (Brlansky et al. 1986; Lee et al. 1987a) . Severe dwarfing of

PAGE 18

5 young trees propagated on sour orange has appeared in many parts of Florida. Large scale outbreaks of induced decline also have appeared in southern Florida, an area previously not affected by CTV. Losses have exceeded 50% in some plantings (Brlansky et al. 1986) . Management of CTV-induced decline in Florida is difficult. The effective use of CTV tolerant rootstocks, such as rough lemon ( Citrus iambhiri Lush.), Troyer citrange { Poncirus trifoliata (L.) Raf. x C. sinensis (L.) Osb.}, Cleopatra mandarin (C. reshni Hort. ex Tanaka) , sweet orange (C. sinensis ) and others (Grant et al. 1961; Wallace, 1978) is diminished by their susceptibility to other diseases, most importantly citrus blight, an endemic disease of unknown etiology which is removing more than 500,000 trees from production annually (Lee et al. 1988a) . The objectives of this research were: i) To evaluate some CTV mild isolates under greenhouse conditions for cross protecting ability against the decline syndrome, and ii) To develop methods for detection of the severe CTV challenge isolate in mixed infections.

PAGE 19

CHAPTER 2 EVALUATION OF THE PROTECTING EFFECTS OF SOME MILD FLORIDA ISOLATES OF CITRUS TRISTEZA VIRUS AGAINST THE DEVELOPMENT OF THE DECLINE SYNDROME Introduction Cross protection is a control strategy used to reduce losses due to plant viral diseases by the use of mild or attenuated strains of a virus which prevent the effect or expression of a related, and usually more severe, strain of the same virus (Fulton, 1986) . Cross protection can be useful when the virus disease is endemic, causes great losses, and no host genetic resistance is available (Fulton, 1986; Gonsalves and Garnsey, 1989: Hamilton, 1985; Muller et al. 1982) . Cross protection has been used commercially to control CTV stem pitting isolates on Pera sweet orange in Brazil (Costa and Muller, 1980; Muller, 1980), and is a part of South Africa's citrus cultivar improvement program to reduce CTVinduced stem pitting on grapefruit (DeLange et al. 1980; Garnsey and Lee, 1988) . Relatively little work has been done to evaluate the potential of cross protection against the CTVinduced decline (CTV-ID) on sour orange, as most countries abandon sour orange as a rootstock when CTV-ID isolates become prevalent. However, experiments conducted in Australia 6

PAGE 20

(Thornton et al. 1980) , United States (Wallace and Drake, 1976; Yokomi et al. 1990) , and Japan (Miyakawa, 1987) indicate that cross protection against CTV-ID isolates may be possible. The classical approach for selecting potential cross protecting mild CTV isolates has been empirical, that is, by collecting a number of CTV isolates from outstanding trees in groves severely affected by the disease (Balaraman and Ramakrishnan, 1980; Muller and Costa, 1987) . Those isolates are propagated on several scion-rootstock combinations in nursery plants and further evaluated over a period of several years under field conditions (Muller, 1980; Muller and Costa, 1977) . This requires the handling and care of large numbers of plants, extensive field space, and considerable time; the results are evaluated over a period of 5-12 years. By this approach, only six mild CTV isolates out of 45 mild isolates originally selected in Brazil were useful for cross protection (Costa and Muller, 1980) . Another approach for the selection of mild CTV isolates for cross protection relies upon host effects for the strain segregation of field collected severe CTV isolates (Roistacher et al. 1988) . The selection of potential cross protecting CTV isolates was made from either symptomless or recovered infected plants after extensive passage of the virus through a series of different citrus and non-citrus hosts in the greenhouse (Roistacher et al. 1987, 1988). This approach provided nine mild or attenuated CTV isolates with outstanding

PAGE 21

8 protection against either seedling yellows or stem pitting CTV isolates from a total of 116 evaluated when challenged with the original isolates from which they were derived. This approach required extensive greenhouse work to follow the segregation of every single isolate, and some attenuated isolates showed a tendency to revert back to their severe forms . Recently, the greenhouse evaluation of CTV-stem pitting isolates as protecting agents against the CTV-ID on sweet/ sour orange combinations was reported (Miyakawa, 1987) . However, CTV-stem pitting isolates as protecting agents offer limited possibilities of commercial use in Florida, particularly because stem pitting isolates still are not present and susceptible species or cultivars such as grapefruit are grown extensively. Citrus tristeza virus causing decline on trees on sour orange rootstock is endemic in Florida. The effective use of CTV-tolerant rootstocks is diminished by their susceptibility to citrus blight. In addition, the increased popularity of sour orange despite tristeza, along with the natural prevalence of mild CTV isolates, provides the opportunity to evaluate cross protection as an alternative control strategy for CTV-ID. However, there are several considerations: i) The threat of recurrent freezes makes it difficult to reliably evaluate the cross protection potential of mild isolates under field conditions, especially in the Ridge area; ii) There is

PAGE 22

9 a need for a method to more rapidly select and evaluate potential cross protecting mild isolates which is adaptable for screening large numbers of isolates; and iii) There is a need to differentiate mild and severe CTV isolates in a mixed infection in the same plant to aid in the evaluation and to better understand the mechanism by which cross protection functions. The objectives of this research were to evaluate naturally occurring Florida mild CTV isolates for cross protecting ability and to develop a methodology to detect the presence of protecting and challenge CTV isolates in mixed infections. The effect of temperature on the cross protecting ability of CTV mild isolates also was studied. Materials and Methods Virus isolates and donor hosts . Five naturally occurring Florida CTV isolates collected from field grown sweet orange or grapefruit trees grafted on sour orange were used in these experiments after transmission by Aphis gossypii Glover. The Tlla, T26, T30 and T55a isolates produce very mild symptoms and little or no stunting on Mexican lime seedlings {Citrus aurantif olia (Christm.) Swingle}, no seedling yellows on Eureka lemon {C. limon (L.) Burm.} or sour orange (C. aurantium L.) and are symptomless in sweet orange {C. sinensis (L.) Osb.} and sweet/sour orange combinations (Garnsey et al. 1987; Lee, 1984; Yokomi and Garnsey, 1987; Yokomi et al. 1987) . The T66a challenge isolate causes strong vein

PAGE 23

10 clearing, stunting and stem pitting in Mexican lime seedlings and severe decline on sweet/sour orange combinations (Garnsey et al. 1987; Yokomi et al. 1987). The CTV isolates were propagated in either C. excelsa Wester, Madam Vinous sweet orange or Mexican lime plants and maintained in a greenhouse with mean minimum and maximum temperatures of 21 and 38°C, respectively. Inoculum source tissue from donor hosts was evaluated by serological indexing (see below) to confirm the presence of CTV before being used as inoculum. Inoculation of CTV isolates and receptor hosts . Oneyear-old Valencia sweet orange plants budded on either sour orange or C. macrophylla Wester rootstocks, were graft inoculated in the stem with each of the CTV mild isolates using three leaf pieces or blind buds per plant (Garnsey and Whidden, 1970; Garnsey et al. 1987). Inoculum tissue was sealed firmly into the receptor stems with plastic grafting tape. Three weeks later the grafting tape was removed and the plants were evaluated for survival of grafted tissue and reinoculated if the grafted tissue had not survived. After verifying by serological indexing that infection by mild isolates had taken place, a minimum of four inoculum pieces of T66a infected tissue were used to challenge the test plants, and they were reinoculated if at least two inoculum pieces were not alive 21 days post-challenge. Surviving inoculum tissue was left in place for the duration of the experiment. Inoculated receptor plants were grown in a

PAGE 24

11 commercial potting mixture (Pro-mix BX) in five liter plastic containers, and fertilized with a mixture of NPK (20-10-20) every other week. Pest and disease management included the application of 0.300 g active ingredient (a. i.) /plant of aldicarb and 0.86 g a.i /L soil drench of ridomil twice a year. The experiment was conducted in a greenhouse with mean minimum and maximum temperatures of 21 and 38°C, respectively. At least 15 plants were inoculated for each CTV isolate/scion/ rootstock combination. A second set of one-year old Valencia sweet orange plants budded on sour orange and C. macrophvlla rootstocks were inoculated with each of the CTV mild isolates and challenged with the T66a severe isolate as previously described. These plants were placed in a greenhouse with controlled mean minimum and maximum temperatures of 21 and 33°C, respectively, to evaluate the effect of temperature on the cross protecting ability of the CTV isolates. At least 10 plants were inoculated per CTV isolate/scion/rootstock combination. Fertilization and plant pest and disease management was as above . Serological tests . CTV infection and relative antigen titer of inoculated plants were determined throughout the study by the double antibody sandwich enzyme-linked immunosorbent assay (DAS-ELISA) (Bar-Joseph et al. 1979b, 1980) , using polyclonal antiserum No. 1053 prepared against whole, unfixed CTV isolate T26 (R.F. Lee, unpublished). The

PAGE 25

12 severe CTV isolate T66a was detected in the challenged plants by DAS-indirect ELISA using the MCA-13 strain specific monoclonal antibody which reacts strongly against most severe CTV isolates (Permar et al. 1990) . For efficiency the serological tests were performed on both experiments at the same time. Routinely, 0.5 g of bark, petioles and midribs of new, fully expanded tissue were finely chopped with a razor blade and ground, using a Tekmar Tissumizer, in 5 ml of phosphate buffered saline (PBS)-Tween + polyvinyl pyrrolidone {PBS = 8 mM Na 2 HP0 4 , 14 mM KH 2 P0 4 , 15 mM NaCl, pH 7.4, (+ 0.1 % Tween 20 + 2% polyvinyl pyrrolidone (PVP-40 Sigma)}. Unless stated otherwise, 200 microliters samples were used per well of the microtiter plates and three washings with PBS-Tween (phosphate buffered saline + 0.1 % Tween 20) were performed between steps. The immunoglobulins (IgG) present in the whole CTV antiserum were purified by the Protein A-Sepharose affinity method (Miller & Stone, 1978) . A portion of purified immunoglobulins were conjugated to alkaline phosphatase by the glutaraldehyde method (Clark et al. 1986). Polystyrene Immulon II microtiter plates (Dynatech Laboratories) were coated with 2.0 ng/ml of purified IgG in carbonate buffer (0.015 M NaHC0 3 , 0.03 M NaC0 3 , pH 9.6) and incubated for 6 hr at 37°C. Antigen samples were added to the wells and incubated for 18 hr at 5°C. Enzyme conjugate was used at a dilution of 1:1,000 in conjugate buffer (PBS-Tween + 2%

PAGE 26

13 polyvinyl pyrrolidone + 0.2% bovine serum albumin) and incubated for 6 hr at 37°C. The reaction with one mg/ml of pnitrophenyl phosphate (Sigma) in 10% triethanolamine, pH 9.8, was measured after 120 min at 405 nm (OD 405 ) with a Labinstruments model EAR 400 AT ELISA plate spectrophotometer. Samples were considered positive when OD 405 values were higher than 0.100 or three times the mean of healthy controls, whichever was greater. For DAS-Indirect ELISA, the microtiter plates were first coated with IgG from antiserum No. 1053. Antigen samples were added as described for DAS-ELI SA. The MCA-13 strain specific monoclonal antibody (hereafter MCA-13) , as ascites fluid, was added at a dilution of 1:5,000 (v/v) in conjugate buffer and incubated 4 hr at 37°C. After washing, goat anti-mouse IgG labeled with alkaline phosphatase (Promega) at a dilution of 1:7,500 (v/v) in conjugate buffer was added and incubated for 2 hr at 37°C. The enzyme reaction was carried out as for DASELISA. For all serological tests, two replications were used per sample. Positive controls included four mild isolates (Tlla, T26, T30, and T55a) and one severe (T66a) CTV isolate. Negative controls included extraction buffer, and similar buffer extracts from healthy C. excelsa and Valencia sweet orange plants. A standard curve prepared with purified CTV T26 isolate diluted to OD 260 values of 0.04, 0.02, 0.01, 0.005, 0.0025, 0.0012, 0.0006, and 0.0003 diluted in buffer extract

PAGE 27

14 of healthy C. excelsa was used to estimate the relative antigen concentration of test samples. Detrimental effects of mild isolates and evaluation of cross protection . To evaluate the effect of each CTV isolate on the inoculated plants and the protecting ability of the mild isolates against the T66a challenge isolate, evaluations were made at five and ten months after the challenge inoculation with the T66a isolate. Phloem necrosis was evaluated by cutting a bark flap at the bud union and the plant tissue was examined with a hand lens for browning. A decline index was assigned for each plant. The parameters scored were stem diameter, plant growth, and foliage symptoms for decline. Each parameter was visually rated from 0 (minimum) to 3 (maximum) , for a maximum cumulative score of 9 for each plant. A high decline index sometimes was accompanied by plant death. The decline index for each treatment was the average of the cumulative scores for all plants in that treatment. Results Antigen titers of mild and severe CTV isolates with polyclonal and MCA-13 monoclonal antibodies . The antigen titers expressed as optical density (OD 405 ) values for the different temperature and host treatments, measured by DASELISA with polyclonal antibodies (PCA) and DAS-indirect ELISA with MCA-13, are summarized in Tables 2.1, 2.2, 2.3, and 2.4. At warm temperatures, Valencia/ sour orange plants inoculated

PAGE 28

15 with mild isolates but unchallenged with T66a gave OD 405 values between 0.091 and 0.145 when analyzed with PCA. The corresponding uninoculated healthy control plants averaged 0.039. The same treatments, including the healthy controls gave values in the range of 0.011-0.025 when analyzed by DASindirect ELISA with MCA-13 . Treatments pre-inoculated with mild isolates and further challenged with T66a gave OD 405 values in the range of 0.130-0.217 with PCA and 0.145-0.189 with MCA-13. The control plants uninoculated with mild isolates but challenged with T66a gave values of 0.174 with PCA and 0.214 with MCA-13 (Table 2.1). Also at warm temperatures, the Valencia/macrophylla plants inoculated with the mild isolates and unchallenged with T66a, gave OD 405 values between 0.092 and 0.164 with PCA. The value for the corresponding uninoculated healthy control plants was 0.040. The same treatments, including the healthy controls gave values in the range of 0.018-0.037 when analyzed with MCA-13. Treatments pre-inoculated with mild isolates and challenged with the T66a gave values in the range of 0.1850.311 with PCA and 0.104-0.305 with MCA-13. The control plants uninoculated with mild isolates but challenged with T66a gave values of 0.194 with PCA and 0.251 with MCA-13 (Table 2.2). The T30 isolate was not evaluated in this portion of the experiment because not enough plants were available. The plants pre-inoculated with the Tlla isolate were not protected and declined and died before the

PAGE 29

16 Table 2 . 1 Relative antigen titer of citrus tristeza virus (CTV) mild isolates in plants unchallenged and challenged by the T66a severe isolate: I. Warm temperature, Valencia/ sour orange . Unchallenged v Challenged 1/2/ CTV OP, 3/ '405 or control plants uninoculated with mild isolate OD 405isolate Polyclonal MCA13 Polyclonal MCA13 Tlla O.lOl^ab^ 0.024 a 0.217 a 0.175 a T26 0.123 ab 0.025 a 0.130 a 0.145 a T30 0.091 ab 0.018 ab 0.212 a 0.189 a T55a 0.145 a 0.014 b 0.194 a 0.181 a Healthy 0.039 b 0.011 b 0.174 a 0.214 a One-year-old plants were graft inoculated with leaf pieces under bark flaps on the stem from donor plants infected with the indicated mild CTV isolates. After verifying virus infection with mild isolates by DAS-ELISA with polyclonal antibodies, the challenged plants were graft inoculated similarly with the T66a severe isolate. Optical density at 405 nm (OD 405 ) was measured after 120 min of reaction. Serological detection was carried out by DAS-ELISA with polyclonal antisera and DAS-indirect ELISA with MCA-13 severe strain specific monoclonal antibody six months after inoculation. Value is the average of duplicate assays of the corresponding plants in Table 2.5. Numbers in the same column followed by different letters are statistically different by Duncan's test (P < 0.05).

PAGE 30

17 Table 2.2 Relative antigen titer of citrus tristeza virus (CTV) mild isolates in plants unchallenged and challenged by the T66a severe isolate: II. Warm temperature, Valencia/macrophylla. Unchallenged 1/ CTV isolate Tlla T26 T30 T55a on 3/ J;o5Polyclonal 0.092/ a i/ 0.132 a NE^ 0.164 a 0.040 a Healthy or control plants uninoculated with mild isolate MCA13 0.026 a 0.030 a NE 0.037 a 0.018 a Challenged OD405— Polyclonal .6/ 0.185 a NE 0.311 a 0.194 a 1/2/ MCA13 0.104 a NE 0.305 a 0.251 a 1 One-year-old plants were graft inoculated with leaf pieces under bark flaps on the stem from donor plants infected with the indicated mild CTV isolates. y After verifying virus infection with mild isolates by DAS-ELISA with polyclonal antibodies, the challenged plants were graft inoculated similarly with the T66a severe isolate. 57 Optical density at 405 nm (OD 405 ) was measured after 120 min of reaction. y Serological detection was carried out by DAS-ELISA with polyclonal antisera and DAS-indirect ELISA with MCA-13 severe strain specific monoclonal antibody six months after inoculation. Value is the average of duplicate assays of the corresponding plants in Table 2.6. 57 Numbers in the same column followed by different letters are statistically different by Duncan's test (P < 0.05). y = severely diseased plants died before serological evaluation. NE = treatment not evaluated.

PAGE 31

18 serological evaluation was made five months after inoculation (Table 2.2) . The OD 405 values were generally higher for the test plants grown at cooler temperatures. The Valencia/sour orange plants inoculated with mild isolates and unchallenged with T66a gave OD 405 values between 0.138 and 0.405 when analyzed with PCA. The corresponding healthy plants averaged 0.039. The same treatments, including healthy controls, gave values in the range of 0.016-0.078 with MCA-13. Treatments pre-inoculated with mild isolates and challenged with T66a gave OD 405 values in the range of 0.174-0.548 with PCA and 0.047-0.477 the MCA13. The control plants uninoculated with mild isolates but challenged with T66a gave OD 405 values of 0.292 with PCA and 0.283 with MCA-13 (Table 2.3). Also at cooler temperatures the Valencia/macrophylla plants inoculated with mild isolates and unchallenged with T66a, gave OD 405 values between 0.335 and 0.361 with PCA. The value for the corresponding uninoculated healthy control plants was 0.036. The same treatments, including healthy controls, gave values in the range of 0.013-0.075 when analyzed with MCA-13. Treatments pre-inoculated with mild isolates and challenged with T66a gave values in the range of 0.406-0.456 with PCA and 0.166-0.320 with MCA-13. The control plants uninoculated with mild isolates but challenged with T66a gave values of 0.432 with PCA and 0.421 with MCA-13 (Table 2.4). The Tlla and T30 isolates were not evaluated for

PAGE 32

19 Table 2.3 Relative antigen titer of citrus tristeza virus (CTV) mild isolates in plants unchallenged and challenged by the T66a severe isolate: III. Cool temperature, Valencia/sour orange. Unchallenged ' Challenged CTY OD^ OD^ isolate Polyclonal MCA-13 Polyclonal MCA-13 Tlla 0.404^ 2> 0.060 a 0.548 a 0.477 a T26 0.405 a 0.078 a 0.195 b 0.055 T30 0.138 a 0.054 a 0.174 b 0.047 T55a 0.308 a 0.055 a 0.276 b 0.057 Healthy 0.039 b 0.016 a 0.292 b 0.283 or control plants uninoculated with mild isolate One-year-old plants were graft inoculated with leaf pieces under bark flaps on the stem from donor plants infected with the indicated mild CTV isolates. After verifying virus infection with mild isolates by DAS -ELI SA with polyclonal antibodies, the challenged plants were graft inoculated similarly with the T66a severe isolate. Optical density at 405 nm (OD 405 ) was measured after 120 min of reaction. Serological detection was carried out by DAS-ELISA with polyclonal antisera and DAS-indirect ELISA with MCA-13 severe strain specific monoclonal antibody six months after inoculation. Value is the average of duplicate assays of the corresponding plants in Table 2.7. Numbers in the same column followed by different letters are statistically different by Duncan's test (P < 0.05).

PAGE 33

20 Table 2 . 4 Relative antigen titer of citrus tristeza virus (CTV) mild isolates in plants unchallenged and challenged by the T66a severe isolate: IV. Cool temperature, Valencia/macrophylla . CTV isolate Tlla T26 T30 T55a Unchallenged v OD 3/ 405Polyclonal NE* 7 0.361^ NE 0.335 a 0.036 b Healthy or control plants uninoculated with mild isolate MCA13 NE 0.049 a NE 0.075 a 0.013 b Challenged 1/2/ OD 405Polyclonal NE 0.406 a NE 0.456 a 0.432 a MCA13 NE 0.166 a NE 0.320 a 0.421 a One-year-old plants were graft inoculated with leaf pieces under bark flaps on the stem from donor plants infected with the indicated mild CTV isolates. After verifying virus infection with mild isolates by DAS-ELI SA with polyclonal antibodies, the challenged plants were graft inoculated similarly with the T66a severe isolate. Optical density at 405 nm (OD 405 ) was measured after 120 min of reaction. NE = treatment not evaluated. Serological detection was carried out by DAS-ELISA with polyclonal antisera and DAS-indirect ELISA with MCA-13 severe strain specific monoclonal antibody six months after inoculation. Value is the average of duplicate assays of the corresponding plants in Table 2.8. Numbers in the same column followed by different letters are statistically different by Duncan's test (P < 0.05).

PAGE 34

21 the Valencia/macrophylla combination (Table 2.4) because not enough plants were available. From the standard curve prepared with purified T26 (Fig. 2.1), it was estimated that an OD 405 value of 0.638 was approximately equivalent to 20 nq/ml of CTV antigen, assuming an extinction coefficient of 2.0 (Gonsalves et al. 1978). Therefore, there was an average of 3.1 fig of CTV antigen per every 100 mg of tissue for each 0.100 OD 405 value in the test samples. Detrimental effects of mild isolates and evaluation of cross protection . The protecting effect of mild isolates was evaluated on the basis of their ability to prevent the detrimental effects on stem diameter, plant growth and foliage symptoms caused under greenhouse conditions by the T66a severe decline isolate. The detrimental effects of mild isolates alone also were evaluated. The number of plants and the scores for the decline index established in every treatment are shown in Tables 2.5, 2.6, 2.7, and 2.8. At warm temperatures, Valencia/sour orange plants inoculated with mild isolates and unchallenged with T66a, and the healthy uninoculated controls, gave overall decline index values in the range of 0.0 and 1.5. Whereas, the plants pre-inoculated with mild isolates and challenged with T66a showed higher decline index values of 6.3, 2.6, 4.2, and 5.5 for the Tlla, T2 6, T30, and T55a mild isolates, respectively. The control plants uninoculated with mild isolates but challenged with

PAGE 35

22 0 0.01 0.02 0.03 0.04 0.05 Purified virus (Optical Density 260 nm) Figure 2.1 Plot of purified citrus tristeza virus (CTV) against optical density. Bark of healthy Citrus excelsa (0.5 g) tissue was ground in 5.0 ml of phosphate buffered saline, pH 7.6, + 0.05% Tween + 2% polyvinyl pyrrolidone and mixed with purified CTV T26 isolate to give the desired optical density at 260 nm (OD 260 ) . DAS-ELI SA was performed as described in materials and methods. An extinction coefficient of 2.0 was assumed (Gonsalves et al. 1978) to estimate the relative virus concentration.

PAGE 36

23 0) 0) (0 i O vo vo EH CD > a) -a o n CD -p (0 CD CP C 0) O fd 03 X •H O 3 TS B CP > eh a) « cp M c o S (0 > § Q) -P w «? •H O 10 \ (0 -H O c 0) rH (0 > H p co *H p •H o <*H o p o a) iP (0 g M 0 0) T3 rH rH JQ .-p -H X c W g 4J O P RS rH 0 C XI VO o in (0 (0 -P (N n in CD rH H -rl Eh Eh EH o ft £ 0) X p c o co a (0 rH rH (0 X! rH a) tj c (0 0) o o -H rH g En X P T3 0) -P (0 0 a) •p (0 o •H •a c H 0) X! P 3 X! O P O -H c £ •H -a 4j a) -h r4 Q) (0 > -P C 10 «J -P rH c a id rH U 04 O c o 0 O U u a> . a) a) c -P o ui a) (1) > a) ^. vo rTg O H ft X! CO rH H U A «J W rH I -rH CO g < -H Q (0 -P a) SI 0 5 <+H (0 r< CP H •rH 0 X 33 0) a) c -p c (0 o •H -P H o a a) _. c •rl 0) CT> c (1) r< „ H « * O r ® -H 01 r4 d) 0) -rl . > 13 Q) . 0 >h J3 nj (D-HH P o C U (0 -H p rH X o XI a) 9 ft „ -r| C tn C 73 0) nJ-H a) > -H S -H •Hh c CT» O • (0 0) CD g ^ -O " O 0 c "J o OflfflO (0 (0 (0 •P 4J rH 0 " (0 3 X! < > 2 * r^ " H X CD rH ^ XJ o II 4J 0) o 0) 10 0) TJ H rl CI C X| >h-H X -P CD CD X § S S * 2 c w c S3S2 X c o 10 e ^•rl tfl X CO CO x a) ty g i 0) Q rH CO T3

PAGE 37

24 T66a averaged a decline index of 4.0 (Table 2.5). The lowest number of dead plants in the experiment (0/5) was obtained when T26 was the protecting isolate (Table 2.5). At warm temperatures, Valencia/macrophylla plants inoculated with mild isolates, including the healthy uninoculated controls gave decline indexes in the range of 1.7-2.2. In comparison, the plants pre-inoculated with mild isolates and challenged with T66a, showed higher decline index values in the range of 2.29. The decline index scores for the control plants uninoculated with mild isolates but challenged with T66a averaged 6.8 (Table 2.6). The lowest number of dead plants occurred when the T26 (0/5) and T55a (0/2) were the protecting isolates (Table 2.6). At cool temperatures, the decline index values for Valencia/sour orange plants inoculated with mild isolates were in the range of 1.0-3.2. The index of 3.0 for the healthy uninoculated controls indicated the generally reduced growth rate of plants at cool temperatures. The decline index values for plants pre-inoculated with mild isolates and challenged with T66a ranged from 5.0 to 7.5. The corresponding control plants uninoculated with mild isolates but challenged with T66a averaged a decline index of 7.5 (Table 2.7). The lowest number of dead plants occurred when T26 (0/2) and T55a (1/4) were the protecting isolates (Table 2.7). Also at cool temperatures, the Valencia/macrophylla plants inoculated with T26 and T55a isolates and the

PAGE 38

25 W X! -0 B rH m rH a> O (0 •o H X! O (0 o P rH c a) • (0 -P 0 rH <4-l CM 55 a rd H <4H 10 O -P c • rd 0 H 2 (X 0) -P rd rH o w in o w 55 CN o vo n T3 0) tr c a) 0) c rH -H X rH rH 0) CM CC o *d cn • x: a> c u Q -h S5 in 55 (M H T3 ami 0) •H x O tr rH d) • • C 0 73 CNJ H Q) a> c rH Q -H rH (0 u i-P Xi c -P o rH O (0 0) rH K O Q) -P rd rH O O C -rH a •H w B 4J C X! rd -P rH -H a) -p rd rH o 03 -H 0) 53 c o 03 (0 rH IH M u U 03 Q) U a> •H rH B •rH P <4H id rH T3 a) -p o a> cntH c 0) -H n > 03 -P C rd 03 P C rd rH u cu o c o TJ b o rH u ra a) <4h >i B a) a) c -P o oi id vo VO Eh cd O rC •H X! > P •H >i H rJ rH CO rH •H , b W -H I (0 W < TJ Q a) >i rd rQ rH 01 u a) o -p c (0 «H rH O +j •h nj _ rH rH a) rH X! -p -H 03 -p c rd c o !p o, 0 rr. -H H a) 03 H rH 5 •H rC > 6 CP 0) . C X3 Q> -H -P +J >i cd 0) O d) W H -H o XI -H -p c cd rH a) > rH 0) -P <*H < 0) rH Q) > a) 03 rH X cd a) 0) 3 C 13 10 -H C •rH rH > CJ a) g X) g Q) -H C ty> C -h a -H rH tP O • 4H CJ d) rH TJ W o c id o S -rH P rd rH § O o . P II •P g n 1 XI •0 0) -H c rd a 0 O o o rd 03 cd > 03 03 e -rH P 03 s o 03 0) c p C rd rH a 03 3 o u o 0> -p. n > (U (0 5 X! rH T 1 -P Q) >H rH 0) c -rH rH U 0) xs X! •H X! 0) X! -P " H X! rH O II a) . rH X! x -p 0) TJ c 0) rH 0) Ml Ul (0 a) tp ceo •H -r| +J rH T) ft O id 6 (DOS a vh 03 ml X 0) -O c -rH B B -rH X rd B • X! --P o rd a) II T> 0) P cd P H rd > a) -p o c p C 0) s p cd 0) rH P w 55

PAGE 39

26 73 id 73 Ul £1 -P C o e a) CP C a) O (0 o a) c H -H X rH rH 0) rt 0 73 CU C U a -h . flm| a> •H X 0 rH a) c o -a (D c • (0 O H 55 ft 0) -p (0 H O 01 n CM O to CM CM IT) IT) If) O IT) in CN CM CM o H .in rt CM EH o o CO CM CO 73 a> -P 0) CM H rt -P rH CO 3 rH 0 o O 10 C -H rH •H O c 73 rl fl rH >i-P -H ,C c m e -P O -P rt rH o C XS O in (0 -P co in a) Eh Eh o ft > a) X! p c o (0 ft rt rH ^ !h (0 XI 0) 73 C fl (0 0) u o CU 73 -rl rH ft-H <4H -a a) a) rH -P id o -rt B -rl ui a) -p rt rH o •rt > EH X! -p -H 73 d) -P (0 0) X! •p 3 X! O -P O -H c J* •H -P 0) tr> a> c -P o to rH (0 c o (0 vo Eh -P •rl >1 < id W rH H -H .4 6 W -rl I u CO < 73 Q CD >i (0 X> rH 10 o a) o -p (0 o +J rl 73 rH •H g -p CP 0) rl 0) Ul p c rt ft c o -H -P 0) 73 •rl c CO Ul u 2 Cp Q) . c x! i rt 73 o U A a) -h -P -p < rt CM| Ul H 0) rl a> > CU Ul (0 fl Ul •rl > o cu >nd A „ 0) 0) rt > -H •H rH CT> o • •H -p rt rH s fl o 0) 4J e •p rt a) M 4J rl 0) Ul rt cu 10 C "O p o fl 73 CU rl -P o rl CP P c rt rH ft >i X! 73 0) -r| c rt ft B o o o rt ui rt ui a) e H -P CU e o Ul x CU 73 C •rl CO o -p -p C * CU fttl-H O CU 73 CP 10 a> rt 5 Q) rl > rt a) p (0 X cu 73 C •H CU c X! ^ -P rH „ ui ui cp n c o o CU Q co| X o CU 73 a) c rl'H v-rl X rt 0 P ft * R° (0 II •p rt cu 73

PAGE 40

27 uninoculated healthy controls showed decline index scores of 5.0, 4.3, and 3.0, respectively. The plants pre-inoculated with the T26 and T55a isolates and challenged with T66a showed scores of 4.2 and 3.5, respectively. The uninoculated controls challenged with T66a showed a decline index of 5.4 (Table 2.8). No dead plants were scored in this experiment 10 months after the challenge inoculations (Table 2.8). Phloem necrosis at the bud union was not observed in any treatment at either temperature at five or ten months after the challenge inoculation (not shown) . Discussion In this study four different naturally occurring Florida mild CTV isolates were evaluated for their cross-protecting ability against the development of the CTV-induced decline in two susceptible scion/rootstock combinations. DAS-ELISA with polyclonal antisera was used to determine the total antigen titer in plants inoculated with mild isolates and those also challenged with the severe T66a isolate. The MCA-13 monoclonal antibody was evaluated in DAS-indirect ELISA for guantitation of the T66a severe challenge isolate in mixed infections. There were some limitations in the transmissibility of CTV by leaf piece grafts from the different hosts used to propagate the CTV isolates. At the beginning of the work most of the CTV isolates had been propagated in Citrus excelsa plants, which has been reported as an excellent propagation

PAGE 41

28 (N 01 A •p c o 73 fi rd 0) o 73 H (0 Eh qj u o rH -P IT) rd • • X! O in -P -H -H 73 > c •H 73 a) a) -P X! fd -P rH d x! in 0 -P O -H c £ •H 73 -p a) <*H -P fd o u -p c oi rd P rH C ft 73 rd 0) rH ^ P 0) ft 0 rn r> fd -p c rH fd 73 O 3 rH rH 73 0 0 o O 01 fi C -H r< O rH -H rd M OCT) ^ 3H >1 >i-P -H fi xj G 01 g a) < rd W rH H -H J fi W -H I 01 w < 73 a a) fc -p >i rd X> rH Ul o Q) O -P rd O 4J W i 4-1 •H (l) > O Xt rd rH 01 O a> oi •H -H 73 r4 0) -P CD 09 rH X rd a) d) -p 3 CO C oi -h c fd •H rH u a) >i73 XI ft fi fi X) Q) -H CP C 73 -H rH Cr> o (1) -H c fd „ •H -P rd rH fi 9 o H CP o 01 rd -H 73 01 rd ft-^ w 01 a) CP rd u CD > rd CD XI -p 01 -H X a) 73 C 01 a) fi -rH P d) fi I o 01 n X 0 a) 4J 73 c c rd a) rH c ft-H rH 01 o a XJ >i< XI 4J rH • rd a> a XI II II X a) O 73 c CD CD C X! O > 01 X a) p o c p c CD fi -P rd a) tH -p w 55

PAGE 42

29 host for purification purposes of CTV (Lee, et al. 1987b, 1988b) . However, it had never been used for leaf graft transmission experiments. At least three inoculations with C. excelsa tissue infected with mild isolates were made unsuccessfully, even though the inoculated tissue survived for at least four weeks post-inoculation and was left in place for several months. It was necessary to switch to Madam Vinous sweet orange and/or Mexican lime plants infected with the mild isolates as inoculum sources before a minimum of 3 or 4 plants per treatment were positively infected with mild isolates as determined by DAS -ELI SA. Furthermore, some treatments, mostly Valencia/macrophylla, were not evaluated because of the lack of replications because not enough plants were available. This originated the need to design another separate experiment, described in Chapter 3, to determine the effectiveness of different citrus species as donor hosts for graft transmission of the virus. At warm temperatures, plants inoculated with mild isolates but unchallenged with T66a had relatively low OD 405 values in the range of 0.095-0.145 (Tables 2.1 and 2.2). At cool temperatures, with few exceptions, were commonly higher in the range of 0.300-0.400 (Tables 2.3 and 2.4). At warm temperatures there were some treatments that gave OD 405 values lower than 0.100, which may be interpreted as negative reactions. However, those values were the averages of the OD 405 readings of every treatment. Thus a single plant with a

PAGE 43

very low OD 405 value could cause the whole treatment average to be lower than 0.100. It has been suggested that a high titer in a plant infected with a mild CTV isolate may be a relative estimate of the protecting ability of mild isolates in crossprotection experiments (Koizumi and Kuhara 1984; Lee et al. 1987a) . Some differences were found in the reaction of the MCA13 monoclonal antibody in the different treatments and temperatures evaluated. At warm temperatures plants preinoculated with mild isolates but unchallenged, gave low OD 405 values in the range of the uninoculated control plants (Table 2.1 and 2.2). Likewise, at cool temperatures, the OD 405 values obtained with the MCA-13 with mild isolates were slightly higher than those obtained at warm temperatures (Tables 2.3 and 2.4). This could be interpreted that the MCA-13 monoclonal antibody may react to some extent with mild isolates when they are above a certain titer in the plants. However, the OD 405 values were always lower than 0.100, which was considered a negative reaction. When plants pre-inoculated with mild isolates and further challenged with the T66a severe isolate were analyzed with the MCA-13 in DAS-indirect ELISA (Tables 2.1 and 2.2), the OD 405 values were generally lower than the unprotected challenged control plants, even though the differences usually were not statistically significant (with the exception T55a in Table 2.2) . At cool temperatures, a similar phenomenon was observed

PAGE 44

31 (Tables 2.3 and 2.4). In this situation, the T26 and T30 isolates generally gave the lowest OD 405 values of all treatments evaluated. This suggests that the mild isolates, especially T26 and T30, but also T55a to some degree, prevented or reduced the multiplication of the T66a challenge isolate. Thus these mild isolates were apparently working as cross-protecting agents. These results provide further evidence of the usefulness of the MCA-13 monoclonal antibody to detect the presence of severe CTV isolates in mixed infections. The MCA-13 has been previously used in other studies to evaluate the presence of severe isolates in field cross-protection experiments (RochaPefia et al. 1990; Yokomi et al. 1990) . Determination of the OD 405 readings when the MCA-13 monoclonal antibody is used to detect the presence of the severe isolate in cross-protection experiments provides a measurable parameter to estimate the ability of mild isolates to prevent the establishment of severe challenge isolates in such experiments. Some differences were found in the effects of CTV mild isolates on the inoculated plants and in their ability to prevent detrimental effects caused by the T66a challenge isolate at different temperatures. At warm temperatures, the effect of the CTV mild isolates on performance of both Valencia/ sour orange and Valencia/macrophylla were negligible. The decline indexes for the healthy uninoculated control plants were nearly egual to or greater than those inoculated

PAGE 45

32 only with mild isolates (Table 2.5 and 2.6). In regard to Valencia/sour orange plants pre-inoculated with mild isolates and challenged with the T66a severe isolate, the lowest decline index (2.6) and lowest number of dead plants (0/5) was obtained with the T26 isolate. Whereas, the highest decline index (6.3) and highest number of dead plants (3/8) was obtained with the Tlla isolate. A decline index of 4.0 and 1/6 dead plants were scored with the uninoculated and challenged control plants (Table 2.5). Also at warm temperatures, Valencia/macrophylla plants pre-inoculated with mild isolates and challenged with the T66a severe isolate, the T26 and T55a isolates obtained the lowest decline index (4.2 and 2.2) and lowest number of dead plants (0/5 and 0/2), respectively. At cooler temperatures, there were some differences in the effect of the CTV mild isolates on growth of both Valencia/sour orange and Valencia/macrophylla. The decline indexes for the healthy uninoculated control plants were variable and ranged from values below to above those obtained with plants inoculated only with mild isolates (Table 2.7 and 2.8). In general, there was a remarkable growth reduction effect at cool temperatures even on healthy uninoculated control plants. In this regard Valencia/sour orange plants pre-inoculated with mild isolates and challenged with the T66a severe isolate, the T26 and T55a isolates obtained the lowest decline index (5.5 and 5.0) and lowest number of dead plants

PAGE 46

(0/2 and 1/4) , respectively. Whereas, the highest decline index (7.5) and highest number of dead plants (3/4) were obtained with the Tlla isolate. A decline index of 7.5 and 1/2 dead plants were scored with the uninoculated and challenged control plants (Table 2.7). At cool temperatures, Valencia/macrophylla plants pre-inoculated with mild isolates and challenged with the T66a severe isolate, had decline index scores similar to the plants inoculated only with mild isolates. No dead plants were obtained in this portion of the experiment (Table 2.8). Of all treatments evaluated at both temperatures, the T26 isolate, and also T55a to some degree, obtained the lowest decline index scores and lowest number of dead plants as compared with the uninoculated challenged control plants. This provided further evidence of their cross-protecting effect, especially the T26 isolate against the development of the CTV-ID syndrome. Of special interest was the high decline index scores and number of dead plants in plants pre-inoculated with the Tlla mild isolate and further challenged with the T66a isolate. It seemed that the combination of Tlla and the T66a isolates produced a more severe reaction on the challenged plants than that caused by the T66a isolate alone in the unprotected control plants. The lack of cross-protecting ability of the Tlla isolate has been previously reported (Yokomi et al. 1987).

PAGE 47

The relatively low decline index scores and low occurrence of dead plants at both temperatures in the unprotected control plants challenged with the T66a isolate, indicates that under greenhouse conditions, the use of one single challenge isolate might not be sufficient to obtain an appropriate rate of decline in a short time basis. It is well documented that CTV occurs naturally as mixtures of isolates or strains with diverse biological properties (Garnsey et al. 1987, McClean, 1974). The T66a severe isolate was originally isolated from an infected field source, and subsequently aphid transmitted to avoid contamination with other viruses (Garnsey et al. 1987; Yokomi and Garnsey, 1987). It is a possibility that part of the original decline components from the field source could have been lost in the subsequent aphid transmissions. To overcome this possibility, it may be advisable in the future to use a mixture of several severe isolates as a challenge to enhance the possibility of obtaining an appropriate occurrence of decline under greenhouse conditions. Another alternative could be the use of higher populations of aphids (50 or 100) to obtain a more complete complex of CTV severe isolates from field samples. Cross-protection using mild virus isolates as a strategy to reduce losses due to CTV has been used in Brazil (Costa and Muller, 1980; Muller, 1980), South Africa (DeLange et al. 1980; Garnsey and Lee, 1988), Japan (Ieki, 1989; Koizumi, 1986) , India (Balaraman and Ramakrishnan, 1980) and Australia

PAGE 48

(Cox et al. 1976; Fraser et al. 1968) , against stem-pitting isolates either in orange, grapefruit, and/or acid lime. Several approaches have been reported for the evaluation of mild isolates under greenhouse conditions. However, these approaches have been addressed mostly to the evaluation of the cross-protecting effect of mild isolates against stem pitting and have included only the host reaction of Mexican lime, sweet orange, or grapefruit seedlings (Roistacher et al. 1987, 1988; Van Vuuren and Noll, 1987) . Another approach where the challenge inoculations are made by using insect vectors to screen mild isolates (Yokomi et al. 1987) has not been extensively used. The results of this work provide further evidence that a) the cross-protection against the CTV-induced decline on sweet/ sour orange combinations may be possible; b) the preliminary evaluation of mild isolates under greenhouse conditions can be made in a relatively short time basis, and c) the severe challenge isolate can be detected by using the MCA-13 strain specific monoclonal antibodies. The recent report of Miyakawa (1987) about the feasibility of crossprotection on sweet/ sour orange combinations, supports these conclusions. Considering the relatively high virus titer found at cool temperatures, it is advisable to propagate the donor plants at temperatures in the range of 21-33°C to better guarantee a high percentage of CTV transmission to the receptor plants.

PAGE 49

36 Likewise, a mixture of several severe isolates should be used as the challenge virus source to give a better evaluation of decline symptoms under greenhouse conditions. The methodology described herein offers the following advantages: i) Depending upon space availability, large numbers of mild isolates can be evaluated uniformly in a time period of 18 to 24 months: from six to twelve months to get the one-year-old plants infected with the mild isolates and verification of infection by serology, two months for challenge and ten months for final evaluation; ii) The availability of the MCA-13 monoclonal antibody provides a useful tool to detect the presence of a severe isolate in the challenged plants, and at the same time allows an estimate of the relative ability of mild isolates to prevent the establishment of the severe isolate in the challenged plants; iii) Mild isolates can be evaluated in grenhouses without the risks that represent the threat of recurrent freezes especially in Florida in recent years, the lack of an appropriate natural challenge pressure (vector or severe isolate) , and the effect of some other devastating diseases (i.e. greening or blight) that can hamper the reliable evaluation of cross-protection experiments under field conditions. Some limitations in the methodology can be also visualized. The use of leaf piece grafts is not always highly efficient to transmit CTV from the donor propagation hosts to the receptor test plants (see Chapter 3) . This leaves the A

PAGE 50

possibility that a lack of transmissibility by leaf piece grafts of the CTV challenge isolate, can be interpreted erroneously as protecting effect by mild isolates. On the other hand, the inoculum tissue with the severe isolate is left in place to enhance the probability of graft-transmission in the challenged plants. This would supply a permanent source of the severe isolate against the mild isolates which may provide a stronger challenge pressure than happens under natural conditions. If this occurs, mild isolates with potential protecting ability under natural challenge conditions could be underestimated or overlooked.

PAGE 51

CHAPTER 3 EFFECTIVENESS OF CITRUS SPECIES AS DONOR HOSTS FOR GRAFT TRANSMISSION OF CITRUS TRISTEZA VIRUS Introduction Citrus tristeza virus (CTV) has long been known to be transmitted by budding and by different grafting procedures (Bennett and Costa, 1949; Bar-Joseph et al. 1979a; Bar-Joseph and Lee, 1990) . In 1951, Wallace experimentally transmitted CTV by placing small portions of donor leaf or bark tissue under a flap of bark on receptor plants. By this method, CTV was transmitted from many field sources of sweet orange to Mexican lime receptor plants, and from Mexican lime to healthy sweet orange plants (Wallace, 1951) . Schwartz (1968) transmitted CTV by connecting the distal portion of the leaf of infected plants to a matching proximal part on a leaf of a receptor plant. By this method, CTV was transmitted to 9 of 20 Mexican lime plants, but transmission was obtained only when callus formation occurred between grafted tissues; also, older and dark green leaves were a better source than younger leaves for both callus formation and virus transmission. Cohen (1972) described a method for CTV transmission in citrus by grafting triangular leaf pieces into triangular holes cut in the leaves of receptor plants. He transmitted 38

PAGE 52

39 CTV from Meyer lemon to 25 of 27 Mexican lime and sour orange seedlings when leaf pieces contained midribs. However, the efficiency of transmission decreased more than 50% when grafts did not include the leaf midrib. A modification of Cohen's procedure, called "leaf-disc grafting" (Blue et al. 1976) involved the use of circular leaf pieces 6 mm in diameter cut from the midrib area of a donor plant leaf and placed into a corresponding hole in the receptor plant leaf. The midrib of the donor tissue is aligned with that of the receptor leaf, and grafts are held in place with transparent tape. This method was as successful as bud inoculation for transmitting many CTV isolates to Mexican lime plants from different citrus species, and was more efficient for transmitting mild CTV isolates. The leaf -disc method was used for routine indexing in the citrus budwood certification program in California (Calavan et al. 1978) . Another method involving the use of leaf piece grafts was reported by Garnsey and Whidden (1970) . Rectangular leaf pieces from infected plants were inserted under corresponding rectangular bark flaps cut in the stem of receptor hosts. This procedure has been used widely for many years with CTV and other citrus viruses, and it has been used in the characterization of the biological properties of diverse worldwide collection of CTV isolates (Garnsey et al. 1987). Leaf piece grafts are especially advantageous when large

PAGE 53

40 numbers of plants are to be inoculated with limited sources of inoculum (Garnsey and Whidden, 1970) . During several experiments with CTV in Florida (Chapter 2; Rocha-Pena et al. 1990) large numbers of plants were inoculated by leaf piece grafts with several CTV isolates that were propagated in different citrus hosts. There were notable differences in the efficiency of transmission of some CTV isolates from different donor hosts, and in some cases no transmission was achieved even after repeated inoculations. The objectives of this research were to evaluate the effect of different citrus hosts on the efficiency of graft transmission of CTV, and to determine the relative distribution of the virus in different host tissues. Materials and Methods Virus isolates and donor hosts . Three isolates of CTV, T26, T30, and T66a, were used throughout the study. They have been described previously (Garnsey et al. 1987; Lee, 1984; Yokomi and Garnsey, 1987) . Virus isolates were propagated in Citrus excelsa Wester, Mexican lime {C. aurantifolia (Christm.) Swingle} and Madam Vinous sweet orange {C. sinensis (L.) Osb.} plants, herein referred to as donor hosts, maintained in a greenhouse with mean minimum and maximum temperatures of 21° and 33°C, respectively. Inoculum tissue from donor hosts was evaluated by serological indexing by the double antibody sandwich enzymelinked immunosorbent assay (DAS-ELISA) (see below) to verify the presence of CTV before

PAGE 54

41 being used as inoculum. All donor host plants were known to be CTV infected for at least one year before the study was started. Grafting procedures and receptor hosts . Rectangular leaf and bark pieces of about 3 X 15 nun were cut from donor hosts with a sharp knife and inserted under corresponding bark flaps cut on the stem of one-year-old Madam Vinous sweet orange, Mexican lime, and grapefruit (C. paradisi Macf.) plants, herein referred to as receptor hosts. A portion of the grafted tissue (2-3 mm) was left exposed at the top of bark flaps to monitor tissue survival at 21 days post-inoculation. A minimum of five plants of each receptor host were each inoculated with 4 pieces of either leaf or bark tissue for every donor host/ virus isolate combination tested. Serological indexing by the double antibody sandwich enzymelinked immunosorbent assay ( DAS-ELI SA) (see below) was carried out on receptor hosts at three and five months postinoculation. Inoculated receptor plants were grown in a commercial potting mixture (Pro-mix BX) in three liter plastic containers, and fertilized with a mixture of NPK (20-10-20) every other week, and given disease and pest management as described in Chapter 2 . Virus distribution and antigen concentration in host tissues . Individual Madam Vinous sweet orange and C. excelsa plants infected with CTV isolates T26 or T66a were used to

PAGE 55

study the relative distribution and antigen concentration of the virus in different tissues of the host plant. Bark, petioles, midribs, and leaf blades of four individual branches of each test plant, were assayed individually by DAS-ELISA. At least four replications were assayed for every host/virus isolate combination tested. Purification of CTV . Citrus tristeza virus was purified from tender new tissue of C. excelsa greenhouse grown plants infected with the T26 isolate, by the Driselase method (Garnsey et al. 1981b; Lee et al. 1988b). The final virus preparations were adjusted with 0.05 M Tris buffer to optical density values (OD 260 ) of 0.4 and stored in one ml aliquots at -18°C. Serological tests . The double antibody sandwich enzymelinked immunosorbent assay (DAS-ELISA) (Bar-Joseph et al. 1979b, 1980) was conducted with polyclonal antiserum no. 1053 prepared against whole, unfixed CTV isolate T26 (R.F. Lee, unpublished) . Polystyrene Immulon II microtiter plates (Dynatech Laboratories) were used. Unless otherwise stated, 200 microliters were used per well of the microtiter plates and, three washings with phosphate buffered saline (PBS)-Tween {PBS = 8 mM Na 2 HP0 4 , 14 mM KH 2 P0 4 , 15 mM NaCl, pH 7.4, (+ 0.1 % Tween 20)} were performed between steps. Host tissue (bark, petioles, midribs, etc.) was chopped finely with a razor blade and ground in a Tekmar Tissumizer in extraction buffer (PBSTween + 2% polyvinyl pyrrolidone (PVP-40 Sigma) at a 1:20

PAGE 56

43 (w/v) dilution. Microtiter plates were coated with 2.0 nq/ml of purified CTV specific IgG in carbonate buffer (0.015 M NaHC0 3 , 0.03 M NaC0 3 , pH 9.6) and incubated for 6 hr at 37°C. Antigen samples were added to the wells and incubated for 18 hr at 5°C. CTV specific IgG conjugated to alkaline phosphatase was used at a dilution of 1:1,000 in conjugate buffer (PBS-Tween + 2% PVP + 0.2% bovine serum albumin) and incubated for 4 hr at 37°C (Bar-Joseph et al. 1979b, 1980). The reaction with one mg/ml of p-nitrophenyl phosphate (Sigma) in 10% triethanolamine, pH 9.8, was measured at 120 min at 405 nm (OD 405 ) with a Bio-Tek EL-307 ELISA plate spectrophotometer. Samples were considered positive when OD 405 values were higher than 0.100 or three times the mean of healthy controls, whichever was greater. There were two replications per sample in each microtiter plate. To estimate the relative CTV concentration in test samples, a standard curve prepared by diluting purified CTV T26 to OD 260 values of 0.04, 0.02, 0.01, 0.005, 0.0025, 0.00125, and 0.0006 in a PBS-Tween + PVP buffered extract of bark of healthy Citrus excelsa it was included as a positive control in every test. Negative controls included PBS-Tween + 2% PVP, conjugate buffer, and extract from healthy C. excelsa . Madam Vinous sweet orange, Mexican lime and grapefruit plants. Results Graft transmission of citrus tristeza virus isolates . At 21 days post-inoculation the survival rate of grafted

PAGE 57

tissue in the whole experiment was 83 and 66 percent for leaf and bark pieces, respectively. Overall at least one of the four grafts survived in 92 and 90 percent of the receptor plants inoculated with leaf or bark pieces, respectively. In calculating the percent of virus transmission for each donor/ receptor host/virus isolate combination, only those plants with at least one (of four) surviving inoculum piece were taken into account. Thus, overall there was a greater efficiency of transmission with leaf pieces (89.2%) than with bark pieces (75.6%) for the whole experiment (Table 3.1). There were three plants of 270 in the entire experiment, one Mexican lime and two grapefruit that became infected even though no successful graft was scored 21 days postinoculation. The overall rate of transmission of CTV by graft inoculation for each donor/ receptor host combination is shown in Table 3.2. With C. excelsa as donor host there was 72.4%, 86.9%, and 60.7% transmission to Madam Vinous, Mexican lime and grapefruit, respectively. With Mexican lime as donor host, there was 93.1%, 76.9% and 89.3% transmission to Madam Vinous, Mexican lime and grapefruit, respectively. With Madam Vinous as donor host there was 86.7%, 100%, and 84.6% transmission to Madam Vinous, Mexican lime and grapefruit respectively. The overall average of transmission was 72.5% from C. excelsa , 85.2% from Mexican lime, and 90.6% from Madam Vinous (Table 3.2). Statistical analysis showed significant

PAGE 58

Table 3 . 1 Transmission of citrus tristeza virus by graft inoculation between selected citrus hosts: I. Efficiency of leaf and bark pieces as inoculum. % plants with Inoculum Inoculum at least one % transmission^ 7 tissue survival %/ successful graft leaf 83. O 1 ^ 92.0 89.2 a bark 66.0 b 90.0 75.6 b y Measured at 21 days post-inoculation. 1 Percent transmission to plants with at least one inoculum piece (of four) alive, measured serologically by DAS-ELISA at 3 and 5 months post-inoculation. Number indicates overall transmission for all donor /receptor /virus isolate combinations. ^ A total of 270 plants (135 each) were inoculated with four pieces of either leaf or bark tissue. Number indicates overall survival for all donor /receptor /virus isolate combinations. y Numbers in the same column followed by different letters are statistically different by Duncan's test (P < 0.05).

PAGE 59

46 X CD X If) o < 00 •H (d rd H 4-1 VO CD • • • & o vo CO CO M O -p (0 O X! a) B X U -H (0 id o rH p ft c o CD CO • • • o 0 vo vo o H rb • (N • • VD 3 ov CO TS id s id W rH CD <0 b -P 0 •H o W X rH c 0 CD -rl W an > o b o H •H 1 (0 w H n) CN 4J > O ° > ° * u o VM CD O CD • H c ft 5 o 5 (0 H » O b c (0 rH P 4H O P , in >~\ id ^ m cd > o p cd £ CD -H -P > cm «n O B b -H c B id w p c CD Ul CD U ft CD rH CD 3 rH id > rC o id w P C CD rH CD 1 rQ IT C -r| o o • • 55 X ill C o •H -P id c •rl ! o CD -P id rH o s rl o p ft CD O u u o vh c o -H (0 CO -H B (0 c id M -p id rH a) > o 0) rC -P Ul CD -P id o -H TJ C •H >H CD X B 2 r\ij n| ^1

PAGE 60

47 differences (P < 0.05%) for all donor-receptor host combinations. Likewise, according to Duncan's multiple range comparison test, there were statistical differences between some of the hosts tested (Table 3.2). The rate of transmission for the three different CTV isolates tested with each donor host is shown in Table 3.3. The T26 isolate was transmitted at a rate of 69.0% to 92.8%. The transmission rates of T30 and T66a isolates ranged from 62.5% to 100% and from 71.4% to 96.7%, respectively, from the hosts tested. While there were statistical differences in the rates of transmission for some of the virus isolate/donor host combination, the overall average of transmission showed no significant differences among them (Table 3.3). The overall statistical analysis for percent transmission of the interactions among the different donor/receptor/virus isolate/ inoculum pieces combinations, indicated no significant differences for receptor and virus isolates alone, and for the combinations of donor/ inoculum pieces, receptor/ inoculum pieces, and for donor/ receptor/ virus isolate. However, significant differences (P < 0.05) were found for donor and inoculum pieces alone, and for the interactions between donor /receptor , donor/virus isolate, receptor /virus isolate, and virus isolate/ inoculum pieces. Virus distribution and antigen concentration in host tissues . The relative antigen titer of CTV as measured by DAS-ELISA in each host tissue/virus isolate combination is

PAGE 61

48 P tfl o U o c o Q CD tr (d rd id *i 0) in in > • • < r-> o m CO CO CO tfl 0) 3 0> (0 rd O C C rd •H M CO o n > o • • • o g P on o r> S w Q) XI g rd id •H rH C CO (N rd • • • 0 cn in vo •rH co VO X CD g r> 0) P rd n rd vo o VD rH (N m vo M o EH Eh Eh -H to > -H tfl tfl 0.0 a Xi tfl rd T3 H CM 0) > •H rH id p a) o o rH rH O c o o CD . -H M 6 3 rH U o c -H a) c o p w rd ai fd u o o tfl CD -h Q) VM IH • •H tfl T3 on >1 H rH p rH rd rd c o •H •H xt p g tfl o •H 0 p fd p P tfl tfl ho a) ^H rH rd to tfl a u CD u -P 0) p rH CD rH or P c C 0 CD T3 U • a) rH tfl 4H rH p 4H td q -H rd T3 U rH o a by J. 4-1 c CN tn 0 C •H •H tfl o tfl 0 -H g rH g rH tfl g o • c •H rd c in U •H c o P g g 3 o rH rd rH rH O VI rd tfl 0 U P Ph CD c ai ^ > a) g O tfl id to at (0 P tfl ft 10 CD ft a) a) -P a> X! -P fd u -p 0 (0 •H CD c TS •H c C rH id -H rd (0 u > Vh C rH 0) 3 CD X! XI Q X! 0 g g rd 3 >i 3 W Z XI z cm| m|

PAGE 62

49 illustrated in Table 3.4. The average optical density values at 405 rati (OD 405 ) for bark tissue were 0.221 and 0.349 for the T26 isolate and 0.266 and 0.336 for the T66a isolate in Madam Vinous and C. excelsa , respectively. The OD 405 values found in the other tissues assayed in both hosts for T26 and T66a isolates were in the range of 0.137 and 0.188 and 0.099 and 0.238 for petioles, 0.173 and 0.241 and 0.049 and 0.133 for midribs, and 0.044 and 0.065 and 0.030 for leaf blades, respectively. There were significant statistical differences between C. excelsa and Madam Vinous for bark tissue with the T26 isolate, and for both petioles and midribs with the T66a isolate. The overall analysis showed that the highest OD 405 values in both hosts for both T26 and T66a isolates, were found in bark, followed by petioles and midribs. Leaf blades showed the lowest OD 405 values of all tissues assayed in both hosts and isolates tested. Some differences in the OD 405 values were found between different parts of the same plant, and from one plant to another, in some virus isolate/host combinations; however, the statistical analysis did not show significative differences among them (data not shown). From the standard curve prepared with purified T26 (Fig. 3.1), it was estimated that an OD 405 value of 0.465 was approximately eguivalent to 20 Mg/ml of CTV, assuming an extinction coefficient of 2.0 (Gonsalves et al. 1978). Therefore, the CTV antigen concentration in the test samples (10 mg of

PAGE 63

50 * VD m CM m to o o o O • • • • o o o o (0 Id XI w rd Xt •H vr> o m o CO CM m T) o H H o •H • • • • S o o o o fd a) P H rH o a • • • • Ph O o o o X) id u 0 « c P 0 •H Q CJ S n (0 0J (0 • -p 0) rH to (0 rH 0) •H (0 10 rH CO o 0 X c 0) •H > 0) r< (d P •0 -H rd u S VD CN En id vo VD Eh C o c o IT) O o I H O o to o p o (d p .5 Si , S; t0 rH xi £ •rl ^Ja H C 3 P sS.S-2 >. o P P -H x; X> H H £ 3 10 &i-H 10 c ft) > , C ft) . b 0 ft) H 3 -h X< (d o * r\ i. . P c (d 3 4H O 6 •H P 0 P td o P ft) 0) c o a) to rH Id O X! ft) u * o ~ p c M a) 1 Xt T3 (1) 0 rH rH o • c — o in •iH O P • id c X! I 0 0 OJ P rd VI a. p w 0) p o to to -H C -» i 10 rH rH ft) id .a o P -H p c to •rH -rl P to
PAGE 64

51 Figure 3.1 Plot of purified citrus tristeza virus (CTV) against optical density. Bark of healthy Citrus excelsa (0.25 g) tissue was ground in 5.0 ml of phosphate buffered saline, pH 7.6, + 0.05% Tween + 2% polyvinyl pyrrolidone and mixed with purified CTV T26 isolate to give the desired optical density at 260 nm (OD 260 ) . DAS-ELISA was performed as described in materials and methods. An extinction coefficient of 2.0 was assumed (Gonsalves et al. 1978) to estimate the relative virus concentration.

PAGE 65

52 tissue/200 /il) ranged from an average of 0.2-0.5 fig in leaf blades to 1.9-2.9 /ig in bark tissue. Discussion In this study three CTV isolates were graft-transmitted by using leaf or bark tissue from three citrus donor hosts to three receptor hosts. The simple establishment and survival of grafted tissue in the receptor host was not sufficient to transmit CTV from certain donor hosts. There were a number of instances, 22 of 80, and 12 of 81, respectively, when C. excelsa and Mexican lime were used as donor hosts, where no transmission was achieved even on those receptor plants where at least one grafted tissue piece was still alive 21 days post-inoculation. Similar results were obtained, but to a lesser degree (7 of 75) when Madam Vinous sweet orange was the donor host. Furthermore, some of the receptor plants where no transmission was scored had all four grafted pieces still alive even five months post-inoculation. The overall analysis of the results showed significant differences in the efficiency of the three donor hosts tested to transmit CTV (Table 3.2) . Likewise, differences were found in the rate of transmission for each donor/receptor host combination. For example, C. excelsa showed rates of transmission of 72.4% and 60.7% to Madam Vinous and grapefruit, respectively; whereas, a rate of transmission of 86.7% was obtained to Mexican lime plants. In regard to Mexican lime as donor host, there was a rate of 89.3% and

PAGE 66

53 93.1.7% transmission to grapefruit and Madam Vinous, respectively and a 76.9% rate to Mexican lime. Transmission from Madam Vinous sweet orange was between 84.6 and 100% in all receptors tested. It was surprising that transmission rates between the same species were only 86.7% for Madam Vinous, and 76.9% for Mexican lime (Table 3.2). Madam Vinous sweet orange was the most efficient donor host with the three receptor hosts tested (90.6%), followed by Mexican lime (85.2%). C. excelsa was a poor donor host (72.5%), being relatively efficient only when inoculated to Mexican lime (Table 3.2). Previous studies on the transmission of CTV by grafting procedures have shown that a period of at least ten days contact between grafted tissues is needed to obtain transmission of the virus to the receptor host (Tolba et al. 1976; Yamaguchi and Patpong, 1980). In this study, the survival of grafted tissue was scored 21 days after inoculation, but the inoculated tissue was left in the receptor plants for up to five months. This should have been ample time for contact between the inoculum and receptor cambium to establish a tissue union with a subsequent transmission of CTV. Furthermore, when leaf pieces were used as inoculum, a small portion of the midrib was included in every piece to increase the success of the grafting. The overall rates of successful grafts were about 83% and a 66.7% for leaf and bark pieces, respectively (Table 3.1).

PAGE 67

54 The reason why a low percentage of graft transmission of the virus was found from some donor hosts, and the absence of an expected 100% when the donor-receptor combination was of the same species, is unknown. A possible explanation could be differences in the virus distribution and/or concentration in the donor tissues used as inoculum. Bark tissue contained the highest antigen titer with OD 405 values in the range of 0.221 and 0.349 in both C. excelsa and Madam Vinous with both CTV isolates tested (Table 3.4). These values were, in some instances, more than double those found in petioles and midribs, and at least triple those found in the leaf blade. Even though the statistical analysis did not show significant differences in antigen titer in either different parts of the same plant or from one plant to another, there were some instances where OD 405 values were as low as the healthy controls. This indicates a possible absence of the virus in those tissues and raises the possibility that occasionally the tissue used for graft transmission may be virus-free, with a subseguent failure in the transmission. Other possibilities could be an occasional absence of phloem connections between the donor and receptor tissues with a subseguent absence of movement of the virus across the junction or the reguirement of a minimum of virus particles present in the tissue used as inoculum in order to accomplish the transmission. Citrus tristeza virus is phloemlimited (Bar-Joseph et al. 1979a; Lister and Bar-Joseph, 1981), and is normally found

PAGE 68

55 at higher concentrations in young phloem-rich tissues (Garnsey, et al. 1979; Bar-Joseph, et al. 1979a); however, the serological titer frequently decreases as the tissues reach maturity or when the plants are exposed to warm environments (Garnsey et al. 1981a; Lee et al. 1988c). The OD 405 values obtained in this research were low if compared with those found when DAS-ELISA is used routinely for CTV diagnosis (Bar-Joseph et al. 1979b; Garnsey et al. 1980b); however, this part of the work was addressed to determine the virus titer in the tissues suitable for graft transmission, and young tender tissue sometimes is not a good source of inoculum for leaf piece grafts (personal observations) . The overall analysis of the results obtained indicates that the efficiency of the graft transmission of CTV is conditioned primarily by the donor/receptor host combination, and secondly by the virus isolate involved, but apparently not by the interaction of the three. For example, C. excelsa showed an overall rate of transmission in the range of 72.5% with all receptor hosts tested (Table 3.2), and a similar low pattern between 69% and 76.7% (= 72.8%) was obtained for the three isolates tested (Table 3.3). Likewise, when Madam Vinous was used as the donor host, there was an overall rate of transmission of 90.6% (Table 3.2). A rate of 77.3-100% (= 88.6%) occurred from this host with the three isolates tested (Table 3) . A comparable event was also scored when Mexican lime was the donor host (Table 3.2 and 3.3). The statistical

PAGE 69

significance found for the interactions donor/receptor and donor/virus isolate, and no significance for the interaction of donor/ receptor /virus isolate supports this conclusion. The use of leaf and/or bark pieces for graft transmission of CTV may be advantageous when large numbers of plants are to be inoculated with limited sources of inoculum (Garnsey and Whidden, 1970) . However, in the light of the results of this research, in order to achieve a high level of transmission, the efficiency of the donor host and the donor/receptor host combination should be considered.

PAGE 70

CHAPTER 4 DEVELOPMENT OF A DOT-IMMUNOBINDING ASSAY FOR CITRUS TRISTEZA VIRUS Introduction Several serological methods have been developed and used to diagnose the presence of citrus tristeza virus (CTV) in infected tissue. These methods include: SDS-immunodif fusion procedures (Garnsey et al. 1979; Bar-Joseph et al. 1980), in situ immunofluorescence (Brlansky et al. 1984; Tsuchizaki et al. 1978), serologically specific electron microscopy (Brlansky et al. 1984; Garnsey et al. 1980), gold immunolabeling microscopy (Davis and Brlansky, 1990) , and several enzyme-linked immunosorbent (ELISA) procedures (BarJoseph and Malkinson, 1980; Bar-Joseph et al. 1979b) including the use of the biotin-avidin system (Irey et al. 1988) and the enzyme-amplified ELISA (Ben-Ze'ev et al. 1988) to increase the sensitivity of detection. Each of these methods has different advantages and sensitivity levels, and therefore, has been used for different purposes and applications (Rocha-Pena and Lee, 1990) . Polyclonal antisera specific for CTV have been developed in different animal species, such as rabbits (Gonsalves et al. 1978; Tsuchizaki et al. 1978; R.F. Lee, unpublished;), and 57

PAGE 71

58 chickens (Bar-Joseph and Malkinson, 1980; Marco and Gumpf, 1990) . Likewise, several monoclonal antibodies (MCAs) have been developed in mice (Gumpf et al. 1987; Permar et al. 1990; Vela et al. 1986, 1988) . The 3DF1 MCA has been reported to react with a broad spectrum of CTV isolates of different geographical origins and is available commercially (Vela et al. 1986, 1988). The MCA-13 MCA (hereafter MCA-13) has been reported to react specifically with CTV isolates that have severe biological activities, especially isolates causing decline on plants grafted on sour orange rootstock (Garnsey and Permar, 1990; Permar et al. 1990). It has been used in diverse studies for strain discrimination purposes (Irey et al. 1988; Garnsey and Permar 1990; Rocha-Pena et al. 1990; Yokomi et al. 1990). The ELISA test, particularly the double antibody sandwich system (DAS-ELISA) (Bar-Joseph et al. 1979b; 1980) has been the most widely used of all serological methods developed for CTV detection (Bar-Joseph et al. 1989; Garnsey, et al. 1981a; Rocha-Pena and Lee, 1991) . In DAS-ELISA the virus in the test sample is trapped and immobilized selectively by specific antibodies adsorbed on polystyrene microtiter plates. Enzymeconjugated antibodies are then reacted with the trapped virus and detected colorimetrically after adding a suitable substrate (Clark and Adams, 1977; Garnsey and Cambra, 1990). DAS-ELISA is relatively easy to perform and is highly sensitive, but it does require some special equipment, it is

PAGE 72

59 laborious and time-consuming for large scale indexing, and both antibodies and large volumes of buffer are used in each test. A simple and rapid serological method, known as dotimmunobinding assay (DIBA) (Hawkes et al. 1982), has been developed and applied for the detection of several plant viruses (Hibi and Saito, 1985; Powell, 1987). The principles of DIBA are similar to those of ELISA, differing mostly in that the antigen and antibodies are bound to nitrocellulose membranes instead of polystyrene microtiter plates, and that the product of the enzyme reaction at the end of the test is insoluble. The use of nitrocellulose membranes for the serological detection of plant viruses has become popular because of the simplicity of equipment required and lower cost of materials needed. In a preliminary study the DIBA test was as sensitive as DAS-ELISA to detect CTV in both field trees and greenhouse grown plants (Rocha-Pena et al. 1990) ; however, some differences were found in the reactivity of the antibodies used, and some nonspecific reactions occurred in the test. The objective of this research was to adapt DIBA for diagnosis of CTV using different polyclonal and monoclonal antibodies and to compare the sensitivity of DIBA with DASELISA and DAS-indirect ELISA. Materials and Methods Antisera used . Polyclonal antisera numbers 1051, 1052, and 1053 previously prepared in rabbits against undegraded,

PAGE 73

60 and unfixed virus particles of T30, T36, and T26 CTV isolates, respectively (R.F. Lee, unpublished) were used. Immunoglobulins (IgG) were purified from whole sera by the Protein A-Sepharose affinity chromatography method (Miller and Stone, 1978) and adjusted to a final concentration of 1.0 mg/ml (OD 280 = 1.40) in PBS buffer with 0.02% sodium azide and stored at 4°C (Clark et al. 1986). The 3DF1 MCA was a gift from Drs. P. Moreno and M. Cambra, Valencia, Spain, and its preparation was described previously (Vela et al. 1986, 1988) . The MCA-13 that reacts specifically with severe CTV strains (Permar et al. 1990) was a gift from Drs. T.A. Permar and S.M. Garnsey. Goat anti-mouse and goat anti-rabbit IgG conjugated with alkaline phosphatase were purchased from either Boehringer or Promega. The purified polyclonal IgG and 3DF1 MCA were tested at concentrations of 1.0, 0.1, 0.2 or 0.01 /xg/ml. The MCA-13 was used as ascites fluid at a dilution of 1:5,000 (v/v) . Goat anti-species IgG were used at concentrations recommended by the manufacturer. Sample preparation .Bark tissue was peeled from fullyexpanded new flushes of CTV infected and healthy citrus plants. The tissue was finely chopped and homogenized with a Tekmar Tissumizer in the presence of Tris buffered saline (TBS)-Tween (TBS = 0.02 M Tris, 0.5 M NaCl, pH 7.5), plus 0.5 % Tween 20 (TBS-Tween) at 1:10 (w/v) dilution.

PAGE 74

61 Dot-immunobindina assay (DIBA) .The general protocol used for DIBA was as follows: nitrocellulose membranes (Micro Separations, Inc.)/ 0.45 /i pore, were cut at a size of 11 X 7.5 cm and wet in TBS for at least 30 min, blotted on chromatography paper (Whatman No. 1) , and allowed to dry for 5 min before use. Aliquots of 2 /il of test samples were applied to nitrocellulose membranes by using as a guide a template constructed from the rack of a micropipet tip holder box, allowed to dry for at least 10-15 min or stored at room temperature for several days before use. All subsequent incubation steps were performed at room temperature in 25 ml of each solution using polypropylene covers of micropipet tip holder boxes as trays. During the incubation or washing the membranes were agitated gently in a shaker at 50 oscillations per min or agitated by hand. Nitrocellulose membranes, with the test samples, were soaked for 30 min in blocking solutions of either 10% horse serum (v//v) , 3% bovine serum albumin (BSA) (w/v) , 3% gelatin (w/v) , 5% Triton X-100 (v/v) , or 0.5% non-fat dry milk (w/v), all in TBS, and washed twice with 2550 ml TBS-Tween and once with TBS, 2 min each. Then the membranes were incubated with the CTV specific antibodies for either 30 min, one, two, or 18 hr, depending on the IgG used, and washed again as after blocking. The membranes were then incubated for 30 min with the corresponding goat anti-species IgG conjugated with alkaline phosphatase. After washing as before, the membranes were incubated in the substrate

PAGE 75

62 solution. The substrate solution was prepared as follows: 10 mg of nitro blue tetrazolium (NBT) (Sigma) were dissolved in 30 ml of TBS substrate buffer (0.1 M Tris, 0.1 M NaCl, 0.005 M MgCl 2 , pH 9.5); then 5 mg of 5-bromo-4-chloro-3-indoyl phosphate (BCIP) (Sigma) were added and dissolved in the solution. The color reaction was stopped by transferring the membranes to distilled water. DAS-ELISA . The double antibody sandwich enzyme-linked immunosorbent assay (DAS-ELISA) (Bar-Joseph et al. 1979b, 1980) was conducted using polyclonal IgG No. 1053 and polystyrene Immulon II microtiter plates (Dynatech Laboratories) . Unless otherwise stated, 200 microliters were used per well and three washings with PBS-Tween (phosphate buffered saline = 8 mM Na 2 HP0 4 , 14 mM KH 2 P0 4 , 15 mM NaCl, pH 7.4, + 0.05% Tween 20) were performed between steps. Microtiter plates were coated using 3 . 0 /xg/ml of purified CTV specific IgG in carbonate buffer (0.015 M NaHC0 3 , 0.03 M NaC0 3 , pH 9.6) and incubated for 6 hr at 37°C. Antigen samples were added to the wells and incubated for 18 hr at 5°C. CTV specific IgG conjugated with alkaline phosphatase at a dilution of 1:1,000 in conjugate buffer {PBS-Tween + 2% polyvinyl pyrrolidone (=PVP 40,000 MW) (w/v) + 0.2% bovine serum albumin (w/v)} was incubated for 4 hr at 37°C. The reaction with p-nitrophenyl phosphate (Sigma) (1.0 mg/ml in 10% triethanolamine, pH 9.8) was guantified after 60 min at

PAGE 76

63 405 nm (OD 405 ) using a Labinstruments model EAR 400 AT ELISA plate spectrophotometer. DAS-indirect ELISA .For DAS-Indirect ELISA the plates were coated with polyclonal IgG No. 1053, and antigen samples were added as described for DAS-ELISA. The 3DF1 MCA was used at a concentration of 0.2 /xg/ml IgG and MCA-13 ascites fluid was used at a dilution of 1:5,000 (v/v) . Each was diluted in conjugate buffer, added to the appropriate plates and incubated for 4 hr 37°C. After washing, alkaline phosphatase conjugated goat anti-mouse IgG was added at a dilution of 1:7,500 (v/v) in conjugate buffer and incubated for 2 hr at 37°C. The enzyme-substrate reaction was carried out as described for DAS-ELISA. Evaluation . Twelve selected naturally-occurring Florida CTV isolates with different biological properties (Garnsey et al. 1987; Permar et al. 1990; Rocha-Pena and Lee, unpublished) were used to determine the range of reactivity of each antibody. All virus source plants were maintained in a greenhouse with mean minimum and maximum temperatures of 21 and 38°C, respectively. The isolates Tlla, T26, T30, T50a and T55a produce very mild symptoms and little or no stunting in Mexican lime seedlings {Citrus aurantifolia (Christm.) Swingle}, no seedling yellows on sour orange (C. aurantium L.) or grapefruit (C. paradisi Macf.) and no symptoms in sweet orange {C. sinensis (L.) Osb.} and sweet/ sour orange combinations. The isolate T4 causes a moderate or stronger

PAGE 77

64 reaction on Mexican lime than isolates listed above. The isolates T3, T36, T62a, T65a, T66a and T67a cause strong vein clearing, stunting and stem pitting in Mexican lime seedlings, varying degrees of decline and/or growth reduction in sweet/ sour orange combinations, varying degrees of seedling yellow reaction and/or severe stunting in sour orange or grapefruit seedlings, and varying degrees of growth reduction in Madam Vinous sweet orange seedlings. The relative sensitivity limits were compared for detection of CTV by DIBA, DAS-ELISA, and DAS-indirect ELISA. An extract {1:10 (w/v) in TBS-Tween} was made of greenhouse grown C. excelsa plants infected with the CTV isolate T36. A series of two fold dilutions was made with a similar extract of healthy C. excelsa . Negative controls included TBS-Tween, and 1:10 extracts of healthy C. excelsa . Madam Vinous sweet orange, Duncan grapefruit and Mexican lime plants, all in TBSTween . Evaluation of different buffers for sample extraction . An additional experiment was carried out to determine the effects of different extraction buffers on the sensitivity of DIBA, DAS-ELISA, and DAS-indirect ELISA. Three different extraction solutions, Tris buffered saline, phosphate buffered saline, and carbonate buffer (all as described above) , with and without 0.05% Tween 20, were used. Three CTV isolates T26, T62a, and T66a propagated in Madam Vinous sweet orange plants were tested in DIBA, DAS-ELISA, and DAS-indirect ELISA

PAGE 78

65 with polyclonal IgG No. 1053 and the 3DF1 MCA as described before. Polyclonal IgG No. 1053 (1.0 /xg/ml) was incubated for 2 hr at 37°C in 1% BSA, 2% PVP, TBS containing 1:200 (v/v) buffer extract of bark from healthy Madam Vinous sweet orange, prior to use in DIBA. Substrate reaction for DAS-ELI SA and DAS-indirect ELISA was quantified after 60 min. Results Development of the dot-immunobindinq assay . The reactivity of each antibody and level of backqround on the nitrocellulose membranes varied with each IqG used, IgG concentration and incubation time. Reactions were affected by the blocking solutions and buffers used for dilution of the antibodies and the commercial goat anti-species IgG conjugates. Six different solutions were evaluated as blocking agents. TBS alone (Fig. 4.1 A) and 10% horse serum (not shown) did not prevent the nitrocellulose sheets from turning dark; whereas 3% BSA, 3% gelatin, 0.5% non-fat dry milk, and 5% Triton X-100 all gave an acceptably white membrane with the different polyclonal IgGs tested (Fig 4.1 B, C, D, E) . The 3% gelatin blocking solution gave the best contrast between the green color for healthy samples and different intensities of a purple color for CTV infected samples (Fig. 4.1 C) . The use of Triton X-100 as a blocking agent partially removed the green material from the nitrocellulose sheets; however, a

PAGE 79

66 Fig. 4.1 Effect of different blocking solutions on the reaction of polyclonal antibodies no. 1053 in dotimmunobinding assay with citrus tristeza virus (CTV) isolates: A. TBS alone; B. 3% bovine serum albumin (BSA) ; C. 3% gelatin; D. 0.5% non-fat dry milk; E. 5% Triton X-100. Number 1, CTV T66a; 2, CTV T26; 3, buffer extract from healthy sweet orange plants; 4, TBS-Tween. Reaction conditions are described in materials and methods.

PAGE 80

67 slight pink reaction was consistently obtained with sap from healthy plants (Fig. 4.1 E) . Polyclonal IgGs had to be cross absorbed with buffer extracts of healthy tissue to reliably discriminate between healthy and CTV infected samples (Fig. 4.2 A, B) , and reactivity varied with concentration and incubation times. The strongest reactions and best discrimination between CTV positive and negative samples were achieved using 1.0 jug IgG/ml and 30 min incubation. An incubation time of 60 min with 1.0 or 0.1 /itg IgG/ml, or longer (18 hr) even with as low as 0.01 /ig IgG/ml frequently resulted in the occurrence of nonspecific reactions with healthy samples, even when the IgG had been cross absorbed with extracts of healthy plants. sA would be expected, MCAs did not have to be cross absorbed with buffer extracts from healthy plants prior to use. The reactivity of the 3DF1 and MCA-13 MCAs was substantially lower than that obtained with any of the polyclonal IgGs tested. For both 3DF1 (1.0 Mg/ml) and MCA-13 (diluted 1:5,000), the incubation time was extended to 2 hr or longer (18 hr) to achieve an adequate positive reaction with infected samples (Fig. 4.2 C, D) . The reactivity of 3DF1 was always slightly lower than that obtained with either polyclonal IgG or the MCA-13 . In initial experiments with polyclonal IgGs, 1% BSA in TBS plus 2% PVP was chosen as the antibody diluent because it consistently gave a whiter background on the nitrocellulose

PAGE 81

68 Fig. 4.2 Reaction of polyclonal and monoclonal antibodies in dot-immunobinding assay with citrus tristeza virus (CTV) isolates. A. Polyclonal antibodies 1053 (1.0 /xg/ml) not cross-absorbed with buffer extract of healthy plant. B. Polyclonal antibodies 1053 (1.0 Mg/ml) incubated with 1:200 (v/v) buffer extract from healthy plants for 2 hr at 37°C prior to use. C and D, monoclonal 3DF1 (1.0 /xg/ml) and MCA13 (1:5,000 dilution) antibodies not cross-absorbed with healthy extract. Number 1, CTV T66a; 2, CTV T26; 3, buffer extract from healthy sweet orange plants; 4, TBS-Tween. Reaction conditions are described in materials and methods.

PAGE 82

69 and stronger reactivity for positive samples. Either the absence of PVP or BSA concentrations of less than 1% in the antibody diluent frequently resulted in a dark background on the nitrocellulose when 3% BSA was used for blocking. Evaluation . The relative sensitivity level of DIBA was measured using increasing dilutions of a plant extract containing CTV isolate T36. This is shown in Figure 4.3. Polyclonal IgGs Nos. 1051, 1052, and 1053 (1.0 /xg/ml IgG for 30 min) consistently gave a distinct positive reaction with CTV T36 diluted to 1/160. There was a slight but consistently positive reaction at the 1/320 dilution and occasionally at 1/640 (Fig. 4.3 A, B, C) . The reactivity of both 3DF1 and MCA-13 MCAs (1.0 Mg/ml and 1:5,000, respectively) was also in the range of 1/160 and 1/320, but only when the incubation time was extended to 18 hr (Fig. 4.3 D, E) . The relative sensitivities of DAS-ELISA and DAS-indirect ELISA were measured similarly against the diluted extract of CTV T36. These results are summarized in Table 4.1. DASELISA, using polyclonal IgG No. 1053 for both coating and conjugate steps, gave OD 405 value of 0.109 (i.e above 0.100) with the CTV T36 sample diluted to 1/160. This was considered a positive reaction. Thus the OD 405 value of 0.070 at a 1/320 dilution was considered a negative reaction. DAS-indirect ELISA, using polyclonal IgG No. 1053 for coating and 3DF1 MCA as second antibody in the double sandwich, gave a positive and negative reaction at dilutions of 1/320 and 1/640,

PAGE 83

70 A B c 1/10 • • • • 1/40 • 1/80 1/160 1/320 1/640 TBS-Tween Fig. 4.3 Relative sensitivity level of different polyclonal and monoclonal antibodies specific to citrus tristeza virus (CTV) in dot-immunobinding assay. Rows A, B and C, polyclonal antibodies nos. 1051, 1052, and 1053, respectively. Rows D and E, monoclonal antibodies 3DF1 and MCA-13, respectively. Extract of Citrus excelsa greenhouse grown plants infected with CTV T3 6 isolate was prepared in TBS-Tween 1:10 (w/v) and two-fold diluted with extract of healthy plants. Reaction conditions are described in materials and methods.

PAGE 84

71 X! W P 3 > P •H < ° So hq P W O O *j " s -H £ -0 ft c w -H CO S o s5 S p (0 w 7 H Q H fa , Q O w >iH -P (0 •H C > O •H H P O •H O W C c o 3 e a) c > (0 •H p ^ * n •H 10 (C N a> Q) o P H >i W ,Q rH -H (COM EH ftP CO H w p u Q) M •H T3 C *rl to < Q < CO H (J W I to < Q CQ H Q co H I < U + + + + + l ro H in CN CO •>* V£> ro 1 a\ o co in in CTi in co H at oo IX) co CN H O o o o rj o o o o O O o o o + + + + + + I 1 co in co CO o in CM co CN o CO in H CTi r> co co fa y3 in vo CO in o o o p co CN CN o o o o o o I + + + + + 1 CN o in o CN co r* H CO o f» (N CN o IT) in r> H H o O O o o H o O o o o o o o o fa a CO ro in col o + rO 03 rH ro a) CD l P O 0 Eh (0 O O o CO X iH o o O o CN CN (1) > o H CN CO i— I CO VD rH Eh W U -H H H H H H H H H Ul 03 I rtj co a) < T3 a a) +3 01 P 0 X 0) a) CO O 0) < £ i •H H A 0 C rvj 51 P , rH 13 "H 0 C • -P >i-H >1 C h i tj (d O CO o ft< X! a) Q -H P
PAGE 85

M 0) i • 43 Ul-P rH § (1) 0 Xi rj P -H 01 > O T3 X! 0) C P H U P 0 X! (0 P rJ -H p X •H Ul 3 H O C o u c + a) > •H P •H U) O II u o rH O O 0 rH a & O a) u c H ft u o a) Q) > •H P id (0 tr> 3 Ul •H > a) c o p Jh M a) a> > p o (0 CP Q) in 1 * c9 > o o a) •H c XJ * > -5 -p ° P •H O Ul o X! ul <4H o o Ul -H Q) M tji ^ P a) o) S c -o o •rH O TJ 0) . H ^•H HI S| Q 4-> 9i S 2J ft n Q) 1-1 U] W -H c c c > 0) OTj -H r; Ul p £ c rH (d O (d O ^l-H O-rl ffl-P •HH£ O P a Cn f0 a a) -h q) O ~ rd g c IT) O M* -P id >i Q) p OJl

PAGE 86

73 respectively. DAS-indirect ELISA using the MCA-13 as second antibody had a sensitivity level similar to DAS-ELISA , i.e. the dilution end point was 1/160 dilution (Table 4.1). The relative reactivity of the different antibodies for the 12 selected CTV isolates in DIBA, and its comparison with both DAS-ELISA and DAS-indirect ELISA, is illustrated in Figure 4.4 and Table 4.2. Polyclonal IgG No. 1053 (1.0 ng/ml) showed a strong positive reaction with all CTV isolates tested in DIBA (Fig. 4.4, left). Polyclonal IgGs Nos. 1051 and 1052 reacted similarly (not shown) . The results obtained with polyclonal IgG No. 1053 in DAS-ELISA with the CTV isolates were the same as with DIBA (Table 4.2). The 3DF1 MCA (1.0 ^g/ral IgG) reacted moderately with most CTV isolates tested, but no reaction and weak or inconclusive reactions were obtained with the isolates T26 and T66a, respectively, in both DIBA and DAS-indirect ELISA (Fig. 4.4, center and Table 4.2). The severe strain specific MCA-13 (used at a dilution of 1:5,000) gave a distinctly positive reaction only with isolates T3, T36, T65a, T66a, and T67a in DIBA and DASindirect ELISA (Fig. 4.4, right and Table 4.2). An inconclusive or slightly positive reaction was obtained occasionally with the T50a, T55a, T4, and T62a isolates in DIBA when the samples were incubated for longer times (18 hr) . Similar reactions occurred with these particular CTV isolates in the DAS-indirect ELISA test (Table 4.2).

PAGE 87

74 A B A B A B 1 o • 2 % • • 3 • • $ • I 4 • % • • 5 6 1 7 8 # • 1053 3DF1 MCA-13 Fig. 4.4 Reaction of polyclonal antibodies no. 1053 and 3DF1 and MCA-13 monoclonal antibodies in dot-immunobinding assay with twelve selected citrus tristeza virus (CTV) isolates. Row A: 1 = Tlla; 2 = T26; 3 = T30; 4 = T50a; 5 = T55a; 6 = T3 ; 7 = T4; 8 = T36. Row B: 1 = T62a; 2 = T65a; 3 = T66a; 4 = T67a. Also in Row B are extract of healthy plants: 5 = Citrus excelsa ; 6 = Madam Vinous sweet orange; 7 = Duncan grapefruit; 8 = Mexican lime. Reaction conditions are described in materials and methods.

PAGE 88

75 i 4 T 4 T 4| 4+ 4 + 4n _j __j 1 1 L Q Q CT* CN CN in in in o n (T\ U i in (J, CN rH CN CTt CXt H CO co 1 — 1 * — ' i — i ^ — ^ ,—1 j in Q rM >-/ o o o O O o o o O o o o -P D Q) rH -H TS + 1 + + + + + + + + + + C -rH CO co CO CO CN CO in H in 1 H VO in n co H VO CT\ CN <* CN CTi m w ID o VO o in H CN < Q Q n H o H CN cn CN H CN CN CN o CN < H a w < Q + + + + + + + + + + + + rCO CO CN VO vo CO o co O co CN o o t> in in CO in in CN H CN co in in in in CN o H O O O o o o o o o o O co < U s H Q ro in o + i + i + + i + + CN| +J W nj 3 rH >H O •H W co| Q 2 S £ W 2 id fC cd fO rrj rrj rrj H VO o O in vo CN in VD H CN CO in in CO CO VO vo VD vo Eh Eh Eh Eh EH Eh EH Eh EH Eh Eh Eh

PAGE 89

76 I CO < Q A P o X) U o m tO a) p OB (0 73 a) to tO id Q) -P (0 x: p CU M CU m H I 3 s c o ft o p (0 o 0 o -p T) CD n n > r> in o H 5S! a in 73 (1) rH CU X! m rH c 3 3* CO 4J M o c >i Tf O XI -H P c (0 a) p CP -1—1 c o u a t7> a) 10 o e i •H P c (0 P (0 O Cn a) X! P >i XI T3 a) o o to 0) u >1 H | o to ft < -h q O TJ C O (0 tn H < CO a) h X5 H' En W O T3 Ph O X( -H • -P >i c T3 (0 O XI -H P c (0 cu p (0 H n CU a u cu p c •H m o rH 0) X! -P -H 0) tO -p c (0 rH • Q> C-H > rH o u c tP (0 o •H X a) £ rH O CU to o XI c cu 0 rH -P tr>-H O *H u cu tw ft f0 rH CP u id xi a) <*H o c (TJ — rH c o p cu 0) tO tO o c p rH -rH TJ tO P e (0 O TJ (0 (d rH S P X CU (0 10 u cu a) <4H o 1 cu CO C rH (0 O cu w rH +J CU 0) > o cu & to 10 II 3 rH • CO P — •rH T3 a> x! .p -H to 3 rH o c o u c •H cu > •H -p -H 10 o ft II H ft 3 Q) (0 £ O P 0) . u u a> cu > -p (0 CP O ^ rH C H n) " in ret c° > o O Q) •H c X! (0 cu a) . o o rH > * i) -H O a -H i to o rH O C rH -H O B o cu H ft rH 3 ft m o a) o C cu to cu rH ft rH O <*H C o •H P (0 3 rH O CU ^rH 4J -O O (TJ Q} U CP rH "P 0) nice TJ O •H O "O to . cu C rH , n C ^ H rH a) (0 >j — to « cu oocj{ Q -P SJ ° 0 B rn S 9J cu m J-> rH (0 > 03 . rH >1 CU -rH • (0 a) a) H •rH rH -P ft 8rH CU o W 10 -H c c c > 0) O (0 -O -H rj 10 P +3 c rH fO O flj U rH -H o -h (0 ft cu -h cu O rH Xl Oh H I CMl ml

PAGE 90

77 Evaluation of different buffers for sample extraction . The reactions of the polyclonal IgG 1053 and MCA 3DF1 with the different CTV isolate/extraction buffer combinations in DIBA are shown in Figure 4.5. In general, there was a strong positive reaction with all the CTV isolate/ extraction buffer combinations tested, with the strongest reaction occurring when PBS without Tween 20 was used as the extraction buffer (Fig. 4.5) . The presence of Tween 20 in the extraction buffer gave a weaker but relatively more uniform reaction spot with both antibodies. The results of the evaluation of the different extraction buffers by DAS-ELISA and DAS-indirect ELISA are shown in Figure 4.6. In DAS-ELISA the strongest reactions occurred with isolate T62a, followed by T26 and T66a, respectively. The presence or absence of Tween 20 made little difference in DAS-ELISA (Fig. 4.6 left). In DAS-indirect ELISA with the 3DF1 MCA the strongest reactions occurred with isolate T26, followed by T62a and T66a, respectively. The presence of Tween 20 gave slightly stronger reactions in most cases. With PBS, the presence of Tween 20 more than doubled the OD 405 values for isolates T62a and T66a, but caused only a slight increase for isolate T26 (Fig. 4.6 right) Discussion The dot-immunobinding assay was adapted for detection of CTV. This included the testing of different agents for blocking and as diluents for the CTV specific antibodies and

PAGE 91

78 T26 T62a In TBS § • < TBST • • < PBS • O PBST • • 1 Carb * • CarbT • H 2 0 o> I CO 10 CM CM ID CO 3 1053 3DF1 Figure 4 . 5 Evaluation of different extraction buffers on the sensitivity of DIBA with citrus tristeza virus (CTV) isolates T26, T62a and T66a. TBS = Tris buffered saline, TBST = TBS containing 0.05% Tween 20, PBS = phosphate buffered saline, PBST = PBS + 0.05% Tween, Carb = carbonate buffer, CarbT = Carb + 0.05% Tween, HC = healthy control. The IgG of polyclonal antibodies No. 1053 or monoclonal antibody 3DF1 were used as primary antibodies followed by the goat antirabbit and anti-mouse IgG conjugates, respectively. Samples were 2 /xl of buffer extracts of greenhouse-grown Madam Vinous sweet orange plants infected with the CTV isolates ground at a 1:10 dilution.

PAGE 92

79 < CO i 11 i i i o o o o o o o 1 1 Q CO <0 KO 1 T I 1 1 ' o O O O O O O W Eh I CO xS < c a to « r! cm c o H -p rd 3 > a) <*> 4-> in (0 o si o + II u CO li CQ a, Eh X) <0 (0 C a) w Q) X! P CN U c ^ a) a) a> Eh U to o -H fO -P N O CD ^ G ccj -HO •H II P f> c £ O "3 O u O H CTfe H Q m O CO H O w co I -P -H 0) +J p c -p . 5 < f0 CO ^3 ro m c ° -P O -a o w p o a) •H Ti C -H I CO < Q O CU CO d o e -rH P c a! P ro o tr> cu x -p >i XI CD o rH rH o 4h c > Eh •H CQ CQ Eh in u CD . CD _ T5 • 2 O -H XI CD XI *0 X5 4J K O I CO o < 0) Eh + CD -P «d 9. 10 ii CO II rrt E" 1 ?> CQ CD ^ O <4H CO (1) -P ro rH a x p T3 CD ^ ^ -H -P CD i) CO rd rH CD (0 P C rd O d o 1-> o c c o o O E CD c rO U O ai P ro tJ> 3 -P T> CD £ CD O > U CO

PAGE 93

80 the commercial goat anti-species IgG conjugates, the evaluation of some CTV specific polyclonal and monoclonal antibodies and the testing of the effect of different extraction buffers with and without Tween 20. In some other systems where nitrocellulose membranes have been used for the immunological detection of proteins either by DIBA (Hibi and Saito, 1985; Powell, 1987; Graddon and Randies, 1986) or by western blotting (Spinola and Cannon, 1985) , some differences have been reported in the suitability of different agents and buffers used for membrane blocking or as diluents for both the virus specific antibodies and the commercial goat anti-species IgG conjugates. In this study, TBS, 10% horse serum, 3% BSA, 3% gelatin, 0.5% non-fat dry milk and 5% Triton X-100 were evaluated as blocking agents. The latter four were found to give an adequately white background on the nitrocellulose membranes to permit discrimination between infected and healthy samples. However, 3% gelatin was used routinely as it gave the most suitable contrast between a green color for the healthy samples and a distinct purple color for the infected samples. It has been well documented that the presence of albumin (BSA or egg ovalbumin) enhances the reactivity of antibodies in diverse serological tests (Clark et al. 1986; Purcifull and Batchelor, 1977) . Likewise, PVP has been commonly used in both extraction and conjugate buffers to prevent non-specific reactions (Clark et al. 1986) . In this study, a concentration

PAGE 94

81 of BSA of less than 1% or the absence of PVP in the antibody diluent frequently resulted in the nitrocellulose membrane turning dark when BSA was used for blocking. Thus TBS containing 1% BSA and 2% PVP was always used as the antibody diluent. Differences were found in the reactivities of the antibodies tested. All polyclonal IgGs (Nos. 1051, 1052, and 1053) reacted similarly in DIBA. After 30 min incubation (at 1.0 /xl/ml IgG) they gave strong positive reactions with CTVinfected samples, but none with healthy samples. However, incubations of 60 min or more sometimes resulted in the occurrence of nonspecific reactions with extracts of healthy plants. This was prevented when the polyclonal antibodies were cross-absorbed with extracts of healthy plants before use. These antibodies gave a strong positive reaction with the 12 selected CTV isolates tested, and had a sensitivity (highest dilution which gave positive reaction) limit of 1/320 with CTV T36 infected samples. The 3DF1 MCA when used at an IgG concentration of 1.0 Hg/ml reacted moderately with most of the CTV isolates tested and had a relative sensitivity limit of 1/320 in DIBA and 1/680 in DAS-indirect ELISA. However, in some cases it was necessary to extend the incubation time (15-18 hr) with 3DF1 in DIBA, because it had a reactivity level that was somewhat lower than any of the polyclonal antibodies tested.

PAGE 95

82 The 3DF1 MCA has been reported to react with a broad spectrum of CTV isolates primarily on the basis of DASindirect ELISA (Vela et al. 1986, 1988) . In this work the 3DF1 reacted weakly with CTV isolates T26 and T66a in some tests (Fig. 4.4, center, A2 and B3) and moderately in others (Fig. 4.2, CI, 2, 5, 6). The reason for this is not known. For isolate T26 it could be due to a low virus concentration as was indicated by DAS-ELISA (Table 4.2). However, the T66a isolate was at a high concentration as indicated in DAS-ELISA with polyclonal antibody 1053 and in DAS-indirect ELISA with MCA-13 (Table 4.2). Yet, in DAS-indirect ELISA 3DF1 gave a low OD 405 value for T66a (Table 4.2). This may indicate a differential reactivity of the 3DF1 MCA with some particular CTV isolates such as T66a. The sensitivity limit for MCA-13 in both DIBA and DASindirect ELISA was near 1/320 (Fig. 4.3 and Table 4.1). In DIBA MCA-13 reacted strongly with CTV isolates T3, T36, T65a, T66a, and T67a, all of which share severe biological properties (Garnsey et al. 1987; Permar, et al. 1990; RochaPena and Lee, unpublished) . However, there were slight positive reaction with CTV isolates T50a and T55a which have mild biological properties and with isolate T62a which has severe properties (Rocha-Pena and Lee, unpublished) (Table 4.2) . The MCA-13 has been reported to react specifically with CTV isolates that have severe biological properties, such as decline, stem pitting, seedling yellows, etc. (Permar and

PAGE 96

83 Garnsey, 1990; Permar et al. 1990), and it has been used for strain discrimination of CTV-induced decline in several field surveys (Irey, et al. 1988; Rocha-Pefia et al. 1990; Yokomi et al. 1990) . The slight positive reaction found frequently in DIBA with some CTV isolates with mild biological properties occurred when the ascites fluid at a dilution of 1:5,000 was incubated for 18 hr. These slight positive reactions could be prevented by using a shorter (2 hr) incubation time or by using the ascites fluid at a dilution of 1:20,000 and an incubation time of 18 hr (Rocha-Peha et al. 1990) . The isolate T62a has been characterized previously as severe on the basis of a severe vein flecking, leaf cupping and stunting of Mexican lime plants, severe growth reduction on grapefruit and Madam Vinous sweet orange seedlings, and moderate growth reduction on sweet/orange combinations (Rocha-Peha and Lee, unpublished) . The weak reaction found with isolate T62a in DIBA and its low OD 405 when MCA-13 was used in DAS-indirect ELISA (Table 4.2), suggests that some severe CTV isolates may not be recognized by the MCA-13. That T62a was in high concentration in these experiments is verified by DAS -ELISA using polyclonal antibodies and DAS-indirect ELISA using the 3DF1 MCA (Table 4.2) . Three buffers, TBS, PBS and carbonate, with and without 0.05% Tween 20, were evaluated for their effects on the sensitivity of DIBA, DAS-ELISA and DAS-indirect ELISA with two antibodies and three CTV isolates. In DIBA, PBS gave the

PAGE 97

84 strongest reactions, followed by TBS and carbonate. The addition of Tween 20 gave slightly weaker reactions with all buffers, but the spots were more uniformly spread on the nitrocellulose. With both ELISA procedures the presence of Tween 20 in the extraction buffers tended to increase the OD 405 values slightly. However, in DAS-indirect ELISA, the presence of Tween 20 in the PBST more than doubled OD 405 readings over PBS alone for CTV isolates T62a and T66a, but had little effect with isolate T26. This did not occur with TBS or the carbonate buffer. One possible explanation is that the presence of Tween 20 in the PBST caused a conformational change in the coat protein of T62a and T66a exposing more of the epitope for binding with the 3DF1 MCA. This appeared to occur to a much lower degree with isolate T26 in DAS-indirect ELISA and with all three of these isolates reacting with the polyclonal antibodies in DAS-ELISA. These results indicate that PBS might be the best extraction buffer for DIBA, and PBST or TBST might be the best for ELISA. However, it would always be advisable to determine the effects of Tween 20 on the serological methods and antigen/ antibody combinations under investigation. There are several advantages to using DIBA over conventional DAS-ELISA or DAS-indirect ELISA for CTV detection. DIBA was rapid and easy to perform and, it was as sensitive as either ELISA procedure for CTV diagnosis. The entire test could be performed in 2-3 hours using polyclonal

PAGE 98

85 antibodies (slightly longer with monoclonal antibodies) , and minimal laboratory eguipment was needed. The use of small containers such as the polypropylene covers for incubation and washing steps enabled the recovery and re-use of both virus specific antibodies and goat anti-species IgG conjugates for at least four weeks when the solutions were properly preserved with 0.02% sodium azide and stored at 5°C between uses. The template constructed from the rack of a micropipet tip holder box enabled the uniform spacing of samples on the nitrocellulose membranes. As pointed out by Powell (1987) , one disadvantage of DIBA as compared to ELISA procedures is the lack of quantitative measurements. However, with appropriate positive and negative controls, DIBA can be used reliably for routine diagnostic work where no quantitative measurements are needed. When quantitation of CTV antigens is required, ELISA should be used.

PAGE 99

CHAPTER 5 SUMMARY AND CONCLUSIONS Four different naturally occurring Florida mild citrus tristeza virus (CTV) isolates were evaluated under greenhouse conditions at two temperature regimes for their crossprotecting ability against the induced decline by a CTV severe challenge isolate in two susceptible scion/rootstock combinations. DAS-ELISA with polyclonal antisera was used to determine the total antigen titer in plants inoculated with mild isolates and those challenged with the severe isolate. The MCA-13 strain specific monoclonal antibody was successfully used in DAS-indirect ELISA for differential isolate detection and quantitation of the severe challenge isolate in mixed infections. The effectiveness of the graft transmission of CTV was evaluated with three CTV isolates by using leaf and bark tissue from three citrus donor hosts to three receptor hosts. The distribution of the virus in different plant tissues was also studied. A dot-immunobinding assay (DIBA) was adapted for CTV detection. The sensitivity level of DIBA was evaluated using three different polyclonal and two monoclonal antibodies and 86

PAGE 100

87 compared with sensitivities of double antibody sandwich (DAS) ELISA and DAS-indirect ELISA with 12 different CTV isolates. The conclusions of this research are summarized as follows: Cross-Protection of Citrus Tristeza Virus At warm temperatures plants pre-inoculated with mild isolates showed relatively low optical density (OD 405 ) values in the range of 0.091-0.145 in DAS-ELISA with polyclonal antisera; whereas, at cool temperatures the values were commonly in the range of 0.300-0.400. At warm temperatures, the MCA-13 monoclonal antibody in plants pre-inoculated with mild isolates but unchallenged, gave OD 405 values in the range of 0.030-0.045 which were as low as the uninoculated control plants. However, at cool temperatures, the OD 405 values obtained with mild isolates were slightly higher than those obtained at warm temperatures, indicating that the MCA-13 monoclonal antibody may react to some extent with mild isolates when they are above a certain titer in the plants. Nevertheless, the OD 405 values were always lower than 0.100, which was considered a negative reaction. When plants pre-inoculated with mild isolates and further challenged with the severe isolate were assayed with the MCA13 monoclonal antibody, they gave typically lower OD 405 values for the T66a isolate than the unprotected challenged control plants which did not have mild isolates. In this regard, the T26 and T3 0 isolates gave the more uniform lower OD 405 values for the T66a isolate in all evaluated treatments. This would

PAGE 101

88 indicate that the mild isolates, especially the T26, T30, and T55a, prevented to some extent the multiplication of the challenge severe isolate, and that they were apparently working as cross-protecting agents. The evaluation of the virus titer of the T66a severe isolate with the MCA-13 monoclonal antibody in the treatments pre-inoculated with mild isolates and further challenged provided a measurable parameter to estimate the ability of mild isolates to prevent the establishment the severe isolate. The effect of mild CTV isolates on performance of both Valencia/sour orange and Valencia/macrophylla was variable. At warm temperatures, the effect was negligible. The decline index for the healthy uninoculated control plants was nearly equal to or greater than those inoculated only with mild isolates. However, at cool temperatures there was a remarkable growth reduction effect in all treatments evaluated. The decline indexes for the healthy uninoculated control plants was variable and ranged from values below to above those obtained with plants inoculated only with mild isolates. Of all treatments evaluated at both temperatures, the T26 isolate, and also T55a to some degree, obtained the lowest decline index scores and lowest number of dead plants as compared with the uninoculated challenged control plants. This provided further evidence of their cross-protecting effect, especially the T26 isolate against the development of the CTVID syndrome.

PAGE 102

89 Plants pre-inoculated with the Tlla mild isolate, and further challenged with the T66a severe isolate showed the highest occurrence of decline, reflected in high decline index scores in the whole experiment, indicating apparently that the combination of both Tlla and T66a isolates produced a more severe reaction in the challenged plants even than that caused by the T66a isolate alone in the unprotected control plants. There was a relatively low occurrence of decline at both temperatures in the unprotected control which were challenged, indicating that to better guarantee an appropriate occurrence of decline symptoms under greenhouse conditions, it would be advisable toe use a mixture of more than one severe isolate as the challenge. The results of this work provide further evidence that: a) the cross-protection against the induced decline on sweet/ sour orange combinations is possible, and it can be evaluated preliminarily under greenhouse conditions in a relatively short time period of 18-24 months. This time period estimate includes from six to twelve months to infect the test plants with mild isolates and verification of infection by serology, two months for challenge inoculation with the severe isolate and ten months for final evaluation; b) The severe challenge isolate can be differentially detected from the mild isolates by using the MCA-13 strain specific monoclonal antibody; c) Mild isolates can be evaluated without the risk

PAGE 103

90 of recurrent freezes that can hamper the reliable evaluation of cross-protection experiments under field conditions. Graft Transmission of Citrus Tristeza Virus The efficiency of graft transmission of CTV by leaf or bark pieces was conditioned primarily by the donor /receptor host combination and secondly by the virus isolate involved. For example, C. excelsa showed a low 72.4% and 60.7% rate of transmission to Madam Vinous and grapefruit, respectively; whereas, an 86.7% rate of transmission was obtained to Mexican lime plants. In regard to Mexican lime as donor hosts, there was a rate of transmission of 89.3% and 93.1.7% to grapefruit and Madam Vinous, respectively and a 76.9% rate to Mexican lime. Rate of transmission from Madam Vinous sweet orange was between 84.6 and 100% in all receptors tested. Of the three donor hosts tested, Madam Vinous sweet orange was the most efficient donor host (90.6%), followed by Mexican lime (85.2%). C. excelsa was a poor donor host with an overall rate of transmission of 72.5%, being relatively efficient only when inoculated to Mexican lime. Differences were found in the rate of transmission of the three different CTV isolates with each donor host tested. However, the overall rate of transmission was in the range of 80% for all isolates. Apparently the efficiency of the graft transmission of the virus depended more on the donor host involved than the virus isolate.

PAGE 104

91 There were some differences in the virus concentration in the donor tissues used as inoculum. Bark tissue contained the highest titer with OD 405 values in the range of 0.221 and 0.349 in both Madam Vinous and C. excelsa with both CTV isolates tested. These values were, in some instances, more than double those found in petioles and midribs, and at least triple those found in the leaf blade. While these values were not statistically significant, some differences were found in virus titer from different parts of the same plant and from one plant to another. There were some instances where OD 405 values were as low as the healthy controls indicating a possible absence of the virus in those tissues. This raises the possibility that occasionally the tissue used for graft transmission may be virus-free, with a subseguent failure in the transmission. The use of leaf and/or bark pieces for graft transmission of CTV may be advantageous when large numbers of plants are to be inoculated with limited sources of inoculum, as is the case for studies of cross-protection. However, in order to achieve a high level in the rate of transmission, the efficiency of donor host and the donor/ receptor host combination should be considered. Leaf pieces provide a better source for inoculation than bark tissue. A Dot-Immunobindina Assay for Citrus Tristeza Virus The dot-immunobinding assay (DIBA) was adapted for detection of CTV. Some differences were found in the J

PAGE 105

92 reactivity of each antibody with the different CTV isolates tested. All polyclonal antibodies (nos. 1051, 1052, and 1053) reacted similarly in DIBA, but all had to be cross-absorbed with buffer extracts of healthy plants before use. All polyclonals gave a strong positive reaction with the 12 CTV isolates tested. The dilution end points of extracts from CTV-infected samples were 1/320 when tested against the polyclonal antibodies. The 3DF1 monoclonal antibody, at a concentration of 1.0 /xg/ml, reacted moderately with most of the CTV isolates tested. Dilution end points of extracts from CTV-infected tissue were 1/320 in DIBA and 1/680 in DAS-indirect ELISA. The strongest reactions in DIBA were obtained when the incubation time was extended to 18 hr. The MCA-13 monoclonal antibody used, at a dilution of 1:5,000 in DIBA, reacted strongly with CTV isolates T3, T36, T65a, T66a, and T67a all sharing common severe biological properties. However, there was a slight positive reaction with two CTV isolates, T50a and T55a, having mild biological properties and with one isolate (T62a) known to have severe properties. The highest dilution end point of CTV-infected extracts which reacted with MCA-13 in DIBA and DAS-indirect ELISA was 1/320 using an incubation time of 18 hr. There are several advantages to using DIBA over conventional DAS-ELISA or DAS-indirect ELISA for CTV detection. DIBA was rapid and easy to perform and, it was as

PAGE 106

sensitive as either ELISA procedure evaluated for CTV diagnosis. The entire test could be performed in 2-3 hours using polyclonal antibodies (slightly longer with monoclonal antibodies) , and minimal laboratory eguipment was needed. The use of small containers such as the polypropylene covers for incubation and washing steps enabled the recovery and re-use of both virus specific antibodies and goat anti-species IgG conjugates for at least four weeks when the solutions were properly preserved with 0.02% sodium azide and stored at 5°C. The template constructed from the rack of a micropipet tip holder box enabled the uniform spacing of samples on the nitrocellulose membranes. One disadvantage of DIBA as compared to ELISA procedures is the lack of quantitative measurements. However, with appropriate positive and negative controls, DIBA can be used reliably for routine diagnostic work where no quantitative measurements are needed. When quantitation of CTV antigens is required, ELISA should be used.

PAGE 107

LITERATURE CITED Balaraman, K. and Ramakrishnan, K. 1980. Strain variation and cross protection in citrus tristeza virus. Pages 60-68, in: Calavan, E.C., Garnsey, S.M., and Timmer, L.W. (eds) . Proc. 8th Conf. Intern. Organ. Citrus Virol. Riverside, California. Bar-Joseph, M. and Lee, R.F. 1990. Citrus tristeza virus. Description of Plant Viruses No. 353 (No. 33 revised) . Commonwealth Mycological Institute/ Association of Applied Biologists. Kew Surrey, England. 7 p. Bar-Joseph, M. and Loebenstein, G. 1973. Effects of strain, source plant, and temperature on the transmissibility of citrus tristeza virus by the melon aphid. Phytopathology 63:716-720. Bar-Joseph, M. and Malkinson, M. 1980. Hen yolk as a source of antiviral antibodies in the enzyme-linked immunosorbent assay (ELISA) : a comparison of two plant viruses. J. Virol. Meth. 1:1-5. Bar-Joseph, M. , Garnsey, S.M., and Gonsalves, D. 1979a. The closteroviruses: a distinct group of elongated plant viruses. Adv. Virus Res. 25:93-168. Bar-Joseph, M. , Garnsey, S.M., Gonsalves, D., and Purcifull, D.E. 1980. Detection of citrus tristeza virus. I. Enzymelinked immunosorbent assay (ELISA) and SDSimmunodif fusion methods. Pages 1-8, in: Calavan, E.C., Garnsey, S.M., and Timmer, L.W. (eds). Proc. 8th Conf. Intern. Organ. Citrus Virol. Riverside, California. Bar-Joseph, M. , Garnsey, S.M., Gonsalves, D., Moscovitz, M. , Purcifull, D.E., Clark, M.F. and Loebenstein, G. 1979b. The use of enzyme-linked immunosorbent assay for the detection of citrus tristeza virus. Phytopathology 69:190-194. Bar-Joseph, M. , Gumpf, D.J., Dodds, J. A. , Rosner, J. A. , and Ginzberg, I. 1985. A simple purification method for citrus tristeza virus and estimation of its genome size. Phytopathology 75:195-198. 94

PAGE 108

95 Bar-Joseph, M. , Marcus, R. , and Lee, R.F. 1989. The continuous challenge of citrus tristeza virus control. Ann. Rev. Phytopathol. 27:292-316. Bar-Joseph, M. , Raccah, B. , and Loebenstein, G. 1977. Evaluation of the main variables that affect citrus tristeza virus transmission by aphids. Proc. Intern. Soc. Citriculture 3:958-961. Bar-Joseph, M. , Roistacher, C.N. , and Garnsey, S.M. 1983. Epidemiology and control of citrus tristeza virus. Pages 61-72, in: Plumb, R.T. and Thresh, J.M. (eds) . Plant Virus Epidemiology. Blackwell Sci. Pub. Oxford. Bar-Joseph, M. , Roistacher, C.N., Garnsey, S.M., and Gumpf, D.J. 1981. A review on tristeza, an ongoing threat to citriculture. Proc. Intern. Soc. Citriculture 1:419-423. Beachy, R.N., Abel, P.P., Nelson, R.S., Register, J., Turner, N.,and Fraley, R.T. 1987. Genetic engineering of plants for protection against virus diseases. Pages 151-159, in: Plant Resistance to Viruses. Ciba Foundation Symposium 133. Wiley, Chichester. Bennett, C.W. and Costa, A.S. 1949. Tristeza disease of citrus. J. Agr. Res. 78:207-237. Ben-Ze'ev, I.S., Frank, A., and Bar-Joseph, M. 1988. Sensitive detection of two plant viruses by enzyme-amplified ELISA. Phytoparasitica 16:343-349. Blue, R.L., Roistacher, C.N. , G. Cartia, and Calavan, E.C. 1976. Leaf disc grafting: A rapid indexing method for detection of some citrus viruses. Pages 207-212, in: Calavan, E.C. (ed.). Proc. 7th Conf . Inter. Organ. Citrus Virol. Riverside, California. Brlansky, R.H. 1987. Inclusion bodies produced in Citrus spp. by citrus tristeza virus. Phytophylactica 19: 211-213. Brlansky, R.H. 1988. Other bacterial canker diseases. Page 7, in: Whiteside, J.O., Garnsey, S.M., and Timmer, L.W. (eds) . Compendium of Citrus Diseases. APS Press. 80 p. Brlansky, R.H., Garnsey, S.M., Lee, R.F., and Purcifull, D.E. 1984. Applications of citrus tristeza virus antisera for use in labeled antibody, immuno-electron microscopical, and sodium dodecyl sulphate-immuno-dif fusion tests. Pages 337-342, in: Garnsey, S.M., Timmer, L.W. , and Dodds, A.J. (eds). Proc. 9th Conf. Intern. Organ. Citrus Virol. Riverside, California.

PAGE 109

96 Brlansky, R.H., Pelosi, R.R., Garnsey, S.M., Youtsey, CO., Lee, R.F., Yokomi, R.K. , and Sonoda, R.M. 1986. Tristeza quick decline epidemic in South Florida. Proc. Fla. State Hort. Soc. 99:66-69. Calavan, E.C., Mather, S.M., and McEachern, E.H. 1978. Registration, certification, and indexing of citrus trees. Pages 185-222, in: Reuter, W. , Calavan, E.C., and Carman, E.G. (eds) . The Citrus Industry. Vol IV. Crop Protection. University of California. Riverside, California. Clark, M.F. and Adams, A.M. 1977. Characteristics of the microplate method of enzyme-linked immunosorbent assay for the detection of plant viruses. J. Gen. Virol. 34:475-483. Clark, M.F., Lister, R.M. , and Bar-Joseph, M. 1986. ELISA techniques. Methods in Enzymology 118:742-767. Cohen, M. 1972. A leaf insert graft used for virus transmission in citrus. Pages 282-284, in: Price, W.C. (ed) . Proc. 5th Conf . Intern. Organ. Citrus Virol. Gainesville, Florida. Costa, A.S. and Muller, G.W. 1980. Tristeza control by cross protection: A U.S. -Brazil cooperative success. Plant Disease 64:538-531. Davis, C.L. and Brlansky, R.H. 1991. Rapid detection of citrus tristeza virus in citrus extracts using a gold labelled antibody. J. Elect. Microsc. Tech. (Abstr., in press). DeLange, J.H., Van Vuuren, S.P., and Bredell, G.S. 1981. Groeipuntenting suiwer sitrusklone vir die superplantskema van virusse. Subtropica 2 (5): 11-16. Fulton, R.W. 1986. Practices and precautions in the use of cross protection for plant virus disease control. Ann. Rev. Phytopathol. 24:67-81. Garnsey, S.M. and Cambra, M. 1990. Enzyme linked immunosorbent assays for citrus pathogens. In: Roistacher, C.N. (ed) . F.A.O. Citrus Disease Indexing Bulletin (in press) . Garnsey, S.M. and Jackson, J.L. 1975. A destructive outbreak of tristeza in central Florida. Proc. Fla. State Hort. Soc. 88:65-69.

PAGE 110

97 Gamsey, S.M. and Lee, R.F. 1988. Tristeza. Pages 48-50, in: Whiteside, J.O., Gamsey, S.M., and Timmer, L.W. (eds) . Compendium of Citrus Diseases. APS Press. 80 p. Gamsey, S.M. and Muller, G.W. 1988. Efficiency of mechanical transmission of citrus tristeza virus. Pages 46-54, in: Timmer, L.W., Garnsey, S.M., and Navarro. L. (ed) . Proc. 10th Conf. Intern. Organ. Citrus Virol. Riverside, California. Garnsey, S.M. and Whidden, R. 1970. A rapid technique for making leaf tissue grafts to transmit citrus viruses. Plant Dis. Rep. 54: 907-908. Garnsey, S.M., Bar-Joseph, M. , and Lee, R.F. 1981a. Applications of serological indexing to develop control strategies for citrus tristeza virus. Proc. Intern. Soc. Citriculture 1:448-452. Garnsey, S.M., Christie, R.G., Derrick, K.S., and BarJoseph, M. 1980a. Detection of citrus tristeza virus. II. Light and electron microscopy of inclusions and viral particles. Pages 9-16, in: Calavan, E.C., Garnsey, S.M., and Timmer, L.W. (eds). Proc. 8th Conf. Intern. Organ. Citrus Virol. Riverside, California. Garnsey, S.M., Gonsalves, D. , and Purcifull, D.E. 1977. Mechanical transmission of citrus tristeza virus. Phytopathology 67:965-968. Garnsey, S.M., Gonsalves, D., and Purcifull, D.E. 1979. Rapid diagnosis of citrus tristeza virus infections by sodium dodecyl sulphate-immunodif fusion procedures. Phytopathology 69:88-95. Garnsey. S.M., Gumpf, D.J., Roistacher, C.N. Civerolo, E.L., Lee, R.F., and Yokomi, R.K. 1987. Toward a standardized evaluation of the biological properties of citrus tristeza virus. Phytophylactica 19:151-157. Garnsey, S.M., Lee, R.F., and Brlansky, R.H. 1981b. Preparation and stability of infectious citrus tristeza virus (CTV) . Phytopathology 71:218 (Abstr.). Garnsey, S.M., Lee, R.F., Youtsey, CO., Brlansky, R.H., and Burnett, H.C.. 1980b. A survey for citrus tristeza virus in registered budwood sources commercially propagated on sour orange rootstock in Florida. Proc. Fla. State Hort. Soc. 93:7-9.

PAGE 111

98 Gonsalves, D. and Garnsey, S.M. 1989. Cross-protection techniques for control of plant virus diseases in the tropics. Plant Disease 73:592-597. Gonsalves, D. , D.E. Purcifull, and Garnsey, S.M. 1978. Purification and serology of citrus tristeza virus. Phytopathology 68:553-559. Grant, T.J., Moreira, S., and Salybe, A. A. 1961. Citrus variety reaction to tristeza virus in Brazil when used in various rootstock and scion combinations. Plant Dis. Rep. 45:416-421. Guerri, J., Moreno, P., and Lee, R.F. 1990. Identification of citrus tristeza virus strains by peptide maps of virion coat protein. Phytopathology 80:692-698. Gumpf, D.J., Zheng, G.-Y., Moreno, P., and Diaz, J.M. 1987. Production and evaluation of specific monoclonal antibodies to citrus tristeza virus strains. Phytophylactica 19:159-161. Hamilton, R.I. 1985. Using plant viruses for disease control. HortScience 20:848-852. Hawkes, R. , Niday, E. , and Gordon, J. 1982. A dotimmunobinding assay for monoclonal and other antibodies. Anal. Bioch. 119:142-147. Hibi, T. and Saito, Y. 1985. A dot-immunobinding assay for the detection of tobacco mosaic virus in infected tissues. J. Gen. Virol. 66:1191-1994. Ieki, H. 1989. The use of cross-protection with mild strains of citrus tristeza virus (CTV) to control stem pitting disease of citrus in Japan. Pages 8-14. FFTC Extension Bulletin No. 284. Ieki, H. and Yamada, S. 1980. Inactivation of citrus tristeza virus (CTV) with heat treatment: Heat tolerance and inactivation of CTV on rootstock-scion combinations. Pages 220-224, in: Calavan, E.C., Garnsey, and Timmer, L.W. (eds) . Proc. 8th Conf. Intern. Organ. Citrus Virol. Riverside, California. Irey, M -S«, Permar, T.A. , and Garnsey, S.M. 1988. Identification of severe isolates of citrus tristeza virus in young field plantings by enzyme-linked immunosorbent assay. Proc. Fla. State Hort. Soc. 101:7376.

PAGE 112

99 Koizumi, M. 1986. Citrus tristeza virus: impact and control by preinoculation in Japan. Pages 148-156, in FFTC Book Series No. 33. Koizumi, M. and Kuhara, S. 1984. Protection of preinoculated citrus trees against tristeza virus in relation to the virus concentration detected by ELISA. Pages 41-48, in: Garnsey, S.M., Timmer, L.W., and Dodds, J. A. (eds) . Proc. 9th Conf. Intern. Organ. Citrus Virol. Riverside, California. Lee, R.F. 1984. Use of double stranded RNAs to diagnose citrus tristeza virus strains. Proc. Fla. State Hort. Soc. 97:5356. Lee, R.F., Brlansky, R.H., and Derrick, K.S. 1988a. Recent progress on studies of citrus blight: Where do we go from here. Citrus Industry 69 (2) :24, 26, 29, 32, 34. Lee, R.F., Brlansky, R.H., Garnsey, S.M., and Yokomi, R.K. 1987a. Traits of citrus tristeza virus important for mild strain cross protection of citrus: The Florida approach. Phytophylactica 19:215-218. Lee, R.F., Calvert, L.A. , Nagel, J., and Hubbard, J.M. 1988b. Citrus tristeza virus : Characterization of coat proteins. Phytopathology 78:1221-122. Lee, R.F., Garnsey, S.M., Brlansky, R.H. , and Goheen, A.C. 1987b. A purification procedure for enhancement of citrus tristeza virus yields and its application to other phloem-limited viruses. Phytopathology 77:543-549. Lee, R.F., Garnsey, S.M., Marais, L.J., Moll, J.N. , and Youtsey, CO. 1988c. Distribution of citrus tristeza virus in grapefruit and sweet orange in Florida and South Africa. Pages 33-38, in: Timmer, L.W. , Garnsey, S.M. and Navarro, L. (eds). Proc. . 10th Conf . Intern. Organ. Citrus Virol. Riverside, California. Lister, R.M. and BarJoseph, M. 1981. Closteroviruses . Pages 809-844, in: Kurstak, E. (ed) . Handbook of Plant Virus Infections and Comparative Diagnosis. Elsevier /NorthHolland Biomedical Press. Canada. Marco, G.M. and Gumpf, D.J. 1990. A simple technique for the production of highly specific polyclonal antisera for citrus tristeza virus. In: Proc. 11th Conf. Intern. Organ. Citrus Virol. Riverside, California (in press).

PAGE 113

100 McClean, A.P.D. 1957. Tristeza virus of citrus: Evidence for absence of seed transmission. Plant Disease Reporter 41: 821. McClean, A.P.D. 1974. The tristeza virus complex. Pages 5966, in: Weathers, L.G. and Cohen, M. (eds) . Proc. 6th Conf . Intern. Organ. Citrus Virol. Riverside, California. Miyakawa, T. 1987. Protection against citrus tristeza seedling yellows infection in citrus by pre-inoculation with stem pitting isolates. Phytophylactica 19:193-196. Miller, T.J. and Stone, H.O. 1978. The rapid isolation of ribonuclease-free immunoglobulin G by Protein A-Sepharose affinity chromatography. J. Immunol. Meth. 4:111-125. Muller, G.W. 1980. Use of mild strains of citrus tristeza virus (CTV) to re-establish commercial production of "Pera" sweet orange in Sao Paulo, Brazil. Proc. Fla. State Hort. Soc. 93:62-64. Muller, G. W. and Costa, A.S. 1977. Tristeza control in Brazil by preimmunization with mild strains. Proc. Intern. Soc. Citriculture 3:868-872. Muller, G.W. and Costa, A.S. 1987. Search for outstanding plants in tristeza infected orchards: The best approach to control the disease by preimmunization. Phytohylactica 19:197-198. Muller, G.W. and Garnsey, S.M,. 1984. Susceptibility of citrus varieties, species, citrus relatives, and non-rutaceous plants to slash-cut mechanical inoculation with citrus tristeza virus (CTV). Pages 33-40, in: Garnsey, S.M., Timmer, L.W. , and Dodds, A.J. (eds) Proc. 9th Conf. Intern. Organ. Citrus Virol. Riverside, California. Muller, G.W., Rezende, J. A.M., and Costa, A.S. 1982. Preimmunization: An approach to control plant virus diseases. Proc. 1st Conf. Impact Viral Dis. Rio de Janeiro, Brazil. Norman, G., Price, W.C., Grant, T.J., and Burnett, H. 1961. Ten years of tristeza in Florida. Proc. Fla. State Hort. Soc. 74:107-111. Permar, T.A. and Garnsey, S.M. 1990. Comparison of biological indexing and immunological assays for identifying severe Florida isolates of citrus tristeza virus. In: Proc. 11th Conf. Intern. Organ. Citrus Virol. Riverside, California (in press) .

PAGE 114

101 Permar, T.A. , Garnsey, S.M., Gumpf, D.J., and Lee, R.F. 1990. A monoclonal antibody that discriminates strains of citrus tristeza virus. Phytopathology 80:224-228. Powell, C.A. 1987. Detection of three plant viruses by dotimmunobinding assay. Phytopathology 77:306-309. Purcifull, D.E. and Batchelor, D.L. 1977. Immunodiffusion tests with sodium dodecyl sulfate (SDS) -treated plant virus and plant inclusions. Technical Bulletin 788. Florida Agricultural Experiment Station. 39 p. Rocha-Pefia, M.A. and Lee, R.F. 1991. Serological techniques for detection of citrus tristeza virus. J. Virol. Meth. (in press) . Rocha-Peha, M.A., Lee, R.F., Permar, T.A., Yokomi, R.K. , and Garnsey, S.M. 1990. Use of enzymelinked immunosorbent and dot-immunobinding assays to evaluate two cross protection experiments after challenge with a severe citrus tristeza virus isolate. In: Proc. 11th Conf. Intern. Organ. Citrus Virol. Riverside, California (in press) . Roistacher, C.N. 1981. A blueprint for disaster: Part II. Changes in the transmissibility of seedling yellows. Calif. Citrograph 67:29-32. Roistacher, C.N., Blue, R.L., Nauer, E.M. , and Calavan, E.C. 1974. Suppression of tristeza virus symptoms in Mexican lime seedlings grown at warm temperatures. Plant Dis. Rep. 58:757-760. Roistacher, C.N., Dodds, J. A. , and Bash, J. A. 1987. Means of obtaining and testing protective strains of seedling yellows and stem pitting tristeza virus: A preliminary report. Phytophylactica 19:199-203. Roistacher, C.N., Dodds, J. A., and Bash, J. A. 1988. Cross protection against citrus tristeza seedling yellows and stem pitting viruses by protective isolates developed in greenhouse plants. Pages 91-100, in: Timmer, L.W. , Garnsey, S.M., and Navarro, L. (eds) . Proc. 10th Conf. Intern. Organ. Citrus Virol. Riverside, California. Schwartz, R.E. 1968. Transmission of the tristeza virus by a leaf union method. S. African J. Agr. Sci. 11:617-622. Spinola, S.M. and Cannon, J.G. 1985. Different blocking agents cause variation in the immunologic detection of proteins transferred to nitrocellulose membranes. J. Virol. Meth. 81: 161-165.

PAGE 115

102 Still, P.E., Hunter, T.J., Rocha-Pena, M.A. , Lee, R.F., and Niblett, C.L. 1990. Western blotting as a rapid method for the immunodetection and classification of citrus tristeza virus isolates. Phytopathology 80: (Abstr.) (in press) . Thornton, I.R., Emmett, R.W. and Stubbs, L.L. 1980. A further report on the grapefruit tristeza pre immunization trial at Mildura, Victoria. Pages 51-53, in: Calavan, E.C., Garnsey, S.M., and Timmer, L.W. (eds) . Proc. 8th Conf. Intern. Organ. Citrus Virol. Riverside, California. Tolba, M.A., Ragab, M.M., and Nour-Eldin, F. 1976. Studies on citrus tristeza virus disease. II. Distribution and movement of the causal virus in citrus plants. Pages 6367, in: Calavan, E.C. (ed) . Proc. 7th Conf. Intern. Organ. Citrus Virol. Riverside, California. Tsuchizaki, T. , Sasaki, A., and Saito, Y. 1978. Purification of citrus tristeza virus from diseased citrus fruits and the detection of the virus in citrus tissues by fluorescent antibody technigues. Phytopathology 68:139142. Van Vuuren, S.P. and Moll, J.N. 1987. Glasshouse evaluation of citrus tristeza virus isolates. Phytophylactica 19:219-221. Vela, C, Cambra, M. , Cortes, E. , Moreno, P., Miguet, J.G., Perez de San Roman, C, and Sanz, A. 1986. Production and characterization of monoclonal antibodies specific for citrus tristeza virus and their use in diagnosis. J. Gen. Virol. 67:91-96. Vela, C, Cambra, M. , Sanz, A., and Moreno, P. 1988. Use of specific monoclonal antibodies for diagnosis of citrus tristeza virus. Pages 55-61, in: Timmer, L.W. , Garnsey, S.M., and Navarro, L. (eds). Proc. 10th Conf. Intern. Organ. Citrus Virol. Riverside, California. Wallace, J.W. 1951. Recent developments in the studies of guick decline and related diseases. Phytopathology 41:785-793. Wallace, J.M. 1978. Virus and virus-like diseases. Pages 67184, in: Reuter, W. , Calavan, E.C, and Carman, E.G. (eds) . The Citrus Industry. Vol IV. Crop Protection. University of California. 362 p. Wallace, J.M. and Drake. 1976. Progress report of studies in California on preimmunization against tristeza in budded

PAGE 116

103 citrus. Pages 58-62, in: Calavan, E.C. (ed) . Proc. 7th Conf . Intern. Organ. Citrus Virol. Riverside, California. Yamaguchi, A. and Patpong, P. 1980. Comparison of time requirement for graft transmission of citrus tristeza virus with other fruit tree viruses. Pages 25-27, in: Calavan, E.C, Garnsey, S.M., and Timmer, L.W. (eds) . Proc. 8th Conf. Intern. Organ. Citrus Virol. Riverside, California. Yokomi, R.K. and Garnsey, S.M. 1987. Transmission of citrus tristeza virus by Aphis qossvpii and Aphis citricola in Florida. Phytophylactica 19:169-172 Yokomi, R.K., Garnsey, S.M., Lee, R.F., and Cohen, M. 1987. Use of insect vectors to screen for protecting effects of mild citrus tristeza virus isolates in Florida. Phytophylactica 19:183-185. Yokomi, R.K., Garnsey, S.M., Permar, T.A. , Lee, R.F., and Youtsey, CO. 1990. Natural spread of severe citrus tristeza virus in citrus preinoculated with mild isolates. In: Proc. 11th Conf. Intern. Organ. Citrus Virol. Riverside, California (in press) .

PAGE 117

BIOGRAPHICAL SKETCH Mario Alberto Rocha-Pena was borne in Monterrey, Nuevo Leon, Mexico, on July 30, 1950. He received the degree of Bachelor of Science in microbiology in 1977 from the Universidad Aut6noma de Nuevo Leon, and in 1979 received his Master of Science degree in plant pathology at the Colegio de Postgraduados, Chapingo, Mexico. From 1979 to 1983 he served as a professor in the Department of Plant Pathology at the Colegio Superior de Agricultura Tropical, Tabasco. In 1984 he accepted a position as a research plant pathologist at the Instituto Nacional de Investigaciones Forestales y Agropecuarias (INIFAP) , in his home state of Nuevo Leon, where he was working until December 1986, before he came to United States to pursue the degree of Doctor of Philosophy in plant pathology at the University of Florida. 104

PAGE 118

I certify that I have read this study and that in my opinion, it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the deqree of Doctor of Philosophy. R.F. Lee, Chairman Professor of Plant Patholoqy I certify that I have read this study and that in my opinion, it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the deqree of Doctor of Philosophy. C. L. Niblett, Cochairman Professor of Plant Patholoqy I certify that I have read this study and that in my opinion, it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the deqree of Doctor of Philosophy. D.E. Purcifull Professor of PI Patholoqy I certify that I have read this study and that in my opinion, it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the deqree of Doctor of Philosophy. ' S , V M . Garnsey Professor Patholoqy I certify that I have read this study and that in my opinion, it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the deqree of Doctor of Philosophy. "g. r. ( y&/w; R.K. Yokomi Assistant Professor of Entomoloqy and Nematoloqy

PAGE 119

I certify that I have read this study and that in my opinion, it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the deqree of Doctor of Philosophy. Dean. /jColleqe of/f December 1990 Aqri&ulture u Dean, Graduate School


52
tissue/200 /xl) ranged from an average of 0.2-0.5 ng in leaf
blades to 1.9-2.9 jug in bark tissue.
Discussion
In this study three CTV isolates were graft-transmitted
by using leaf or bark tissue from three citrus donor hosts to
three receptor hosts. The simple establishment and survival
of grafted tissue in the receptor host was not sufficient to
transmit CTV from certain donor hosts. There were a number
of instances, 22 of 80, and 12 of 81, respectively, when C.
excelsa and Mexican lime were used as donor hosts, where no
transmission was achieved even on those receptor plants where
at least one grafted tissue piece was still alive 21 days
post-inoculation. Similar results were obtained, but to a
lesser degree (7 of 75) when Madam Vinous sweet orange was the
donor host. Furthermore, some of the receptor plants where
no transmission was scored had all four grafted pieces still
alive even five months post-inoculation.
The overall analysis of the results showed significant
differences in the efficiency of the three donor hosts tested
to transmit CTV (Table 3.2). Likewise, differences were found
in the rate of transmission for each donor/receptor host
combination. For example, C. excelsa showed rates of
transmission of 72.4% and 60.7% to Madam Vinous and
grapefruit, respectively; whereas, a rate of transmission of
86.7% was obtained to Mexican lime plants. In regard to
Mexican lime as donor host, there was a rate of 89.3% and


61
Dot-immunobindinq assay (PIBA).- The general protocol
used for DIBA was as follows: nitrocellulose membranes (Micro
Separations, Inc.)/ 0.45 m pore, were cut at a size of 11 X
7.5 cm and wet in TBS for at least 30 min, blotted on
chromatography paper (Whatman No. 1), and allowed to dry for
5 min before use. Aliquots of 2 /xl of test samples were
applied to nitrocellulose membranes by using as a guide a
template constructed from the rack of a micropipet tip holder
box, allowed to dry for at least 10-15 min or stored at room
temperature for several days before use. All subsequent
incubation steps were performed at room temperature in 25 ml
of each solution using polypropylene covers of micropipet tip
holder boxes as trays. During the incubation or washing the
membranes were agitated gently in a shaker at 50 oscillations
per min or agitated by hand. Nitrocellulose membranes, with
the test samples, were soaked for 30 min in blocking solutions
of either 10% horse serum (v//v), 3% bovine serum albumin
(BSA) (w/v), 3% gelatin (w/v), 5% Triton X-100 (v/v), or 0.5%
non-fat dry milk (w/v), all in TBS, and washed twice with 25-
50 ml TBS-Tween and once with TBS, 2 min each. Then the
membranes were incubated with the CTV specific antibodies for
either 30 min, one, two, or 18 hr, depending on the IgG used,
and washed again as after blocking. The membranes were then
incubated for 30 min with the corresponding goat anti-species
IgG conjugated with alkaline phosphatase. After washing as
before, the membranes were incubated in the substrate


73
respectively. DAS-indirect ELISA using the MCA-13 as second
antibody had a sensitivity level similar to DAS-ELISA i.e.
the dilution end point was 1/160 dilution (Table 4.1).
The relative reactivity of the different antibodies for
the 12 selected CTV isolates in DIBA, and its comparison with
both DAS-ELISA and DAS-indirect ELISA, is illustrated in
Figure 4.4 and Table 4.2. Polyclonal IgG No. 1053 (1.0 /ng/ml)
showed a strong positive reaction with all CTV isolates tested
in DIBA (Fig. 4.4, left). Polyclonal IgGs Nos. 1051 and 1052
reacted similarly (not shown). The results obtained with
polyclonal IgG No.1053 in DAS-ELISA with the CTV isolates were
the same as with DIBA (Table 4.2).
The 3DF1 MCA (1.0 /xg/ml IgG) reacted moderately with most
CTV isolates tested, but no reaction and weak or inconclusive
reactions were obtained with the isolates T26 and T66a,
respectively, in both DIBA and DAS-indirect ELISA (Fig. 4.4,
center and Table 4.2).
The severe strain specific MCA-13 (used at a dilution of
1:5,000) gave a distinctly positive reaction only with
isolates T3, T36, T65a, T66a, and T67a in DIBA and DAS-
indirect ELISA (Fig. 4.4, right and Table 4.2). An
inconclusive or slightly positive reaction was obtained
occasionally with the T50a, T55a, T4, and T62a isolates in
DIBA when the samples were incubated for longer times (18 hr) .
Similar reactions occurred with these particular CTV isolates
in the DAS-indirect ELISA test (Table 4.2).


51
Purified virus (Optical Density 260 nm)
Figure 3.1 Plot of purified citrus tristeza virus (CTV)
against optical density. Bark of healthy Citrus excelsa (0.25
g) tissue was ground in 5.0 ml of phosphate buffered saline,
pH 7.6, + 0.05% Tween + 2% polyvinyl pyrrolidone and mixed
with purified CTV T2 6 isolate to give the desired optical
density at 260 nm (OD260) DAS-ELISA was performed as described
in materials and methods. An extinction coefficient of 2.0
was assumed (Gonsalves et al. 1978) to estimate the relative
virus concentration.


102
Still, P.E., Hunter, T.J., Rocha-Pea, M.A., Lee, R.F., and
Niblett, C.L. 1990. Western blotting as a rapid method
for the immunodetection and classification of citrus
tristeza virus isolates. Phytopathology 80: (Abstr.) (in
press).
Thornton, I.R., Emmett, R.W. and Stubbs, L.L. 1980. A further
report on the grapefruit tristeza preimmunization trial
at Mildura, Victoria. Pages 51-53, in: Calavan, E.C.,
Garnsey, S.M., and Timmer, L.W. (eds). Proc. 8th Conf.
Intern. Organ. Citrus Virol. Riverside, California.
Tolba, M.A., Ragab, M.M., and Nour-Eldin, F. 1976. Studies on
citrus tristeza virus disease. II. Distribution and
movement of the causal virus in citrus plants. Pages 63-
67, in: Calavan, E.C. (ed). Proc. 7th Conf. Intern.
Organ. Citrus Virol. Riverside, California.
Tsuchizaki, T., Sasaki, A., and Saito, Y. 1978. Purification
of citrus tristeza virus from diseased citrus fruits and
the detection of the virus in citrus tissues by
fluorescent antibody techniques. Phytopathology 68:139-
142.
Van Vuuren, S.P. and Moll, J.N. 1987. Glasshouse evaluation
of citrus tristeza virus isolates. Phytophylactica
19:219-221.
Vela, C., Cambra, M., Corts, E., Moreno, P., Miguet, J.G.,
Prez de San Romn, C., and Sanz, A. 1986. Production and
characterization of monoclonal antibodies specific for
citrus tristeza virus and their use in diagnosis. J. Gen.
Virol. 67:91-96.
Vela, C., Cambra, M., Sanz, A., and Moreno, P. 1988. Use of
specific monoclonal antibodies for diagnosis of citrus
tristeza virus. Pages 55-61, in: Timmer, L.W., Garnsey,
S.M., and Navarro, L. (eds). Proc. 10th Conf. Intern.
Organ. Citrus Virol. Riverside, California.
Wallace, J.W. 1951. Recent developments in the studies of
quick decline and related diseases. Phytopathology
41:785-793.
Wallace, J.M. 1978. Virus and virus-like diseases. Pages 67-
184, in: Reuter, W., Calavan, E.C., and Carman, E.G.
(eds). The Citrus Industry. Vol IV. Crop Protection.
University of California. 362 p.
Wallace, J.M. and Drake. 1976. Progress report of studies in
California on preimmunization against tristeza in budded


4
which is effective against stem pitting (Bar-Joseph et al.
1989; Garnsey and Lee, 1988; Lee et al. 1987a). Genetic
resistance to CTV is not available in commercially acceptable
scions (Bar-Joseph et al. 1989; Garnsey and Lee, 1988). The
application of genetically engineered cross protection,
currently effective in several other crops (Beachy et al.
1987), is an attractive possibility for CTV control in the
future.
CTV has been widespread in Florida for many years and
induced decline has occurred in localized areas (Garnsey and
Jackson, 1975; Norman et al. 1961). However, until recently,
it had not caused major losses because most of the citrus
acreage had been propagated on CTV-tolerant rootstocks and
because of the prevalence of mild CTV isolates which did not
seriously affect trees grafted on sour orange rootstock
(Brlansky et al. 1986; Garnsey et al. 1980b; Lee et al.
1987a) In the last decade the situation in Florida has
changed radically. Sour orange continued to be a very popular
rootstock because of cold tolerance, high fruit quality of the
scion, and its tolerance to citrus blight. The high demand
for plants on sour orange, plus discovery of citrus bacterial
leaf spot in some nurseries (Brlansky, 1988; Garnsey, personal
communication), caused nurserymen to use budwood from source
trees that had not been propagated previously on sour orange.
Many such trees apparently were harboring severe CTV isolates
(Brlansky et al. 1986; Lee et al. 1987a). Severe dwarfing of


C. excelsa
(HC) -


0.021 -
0.033 -
0.013
M. Vinous
(HC) -
-
-
0.004 -
0.045 -
0.000
Grapefruit
(HC) -
-
-
0.004 -
0.014 -
0.008
M. lime
(HC) -
-
-
0.009 -
0.029 -
0.038
J The IgG of polyclonal antibody no. 1053 was used to coat the plates for both DAS-
ELISA and DAS-indirect ELISA. For DAS-ELISA the no. 1053 IgG conjugate was used as
second antibody. For DAS-indirect ELISA the unlabeled 3DF1 and MCA-13 were the
intermediate antibodies followed by the goat anti-mouse IgG conjugate.
2 / ,
' Buffer extracts (1:10 dilution) of bark from greenhouse grown plants either of C.
excelsa. Madam Vinous sweet orange, grapefruit or Mexican lime.
3 / ,
' M = mild; MD = moderate; S = severe; HC = healthy control; on the basis of symptom
reaction on a series of citrus hosts (Garnsey et al. 1987; Permar et al. 1990; Rocha-
Pea and Lee, unpublished).
/ Visual evaluation for presence of a purple color (+ = positive, +/- = inconclusive, -
= negative).
Optical density at 405 nm (D405) after 60 min of substrate reaction. Mean of two
replications per plate. Reactions were considered positive (+) when OD4Q5 values were
higher than three times the mean of healthy controls or 0.100, whichever was greater.
Reactions with lower values were considered negative (-).


Optical Density
DAS-ELISA
DAS-indirect ELISA
Figure 4.6 Evaluation of different extraction buffers on the sensitivity of DAS-ELISA
and DAS-indirect ELISA with citrus tristeza virus (CTV) isolates T26, T62a and T66a.
TBS = Tris buffered saline, TBST = TBS containing 0.05% Tween 20, PBS = phosphate
buffered saline, PBST = PBS + 0.05% Tween, Carb = carbonate buffer, CarbT = Carb + 0.05%
Tween, HC = healthy control. The IgG of polyclonal antibodies No. 1053 was used to coat
the plates for both DAS-ELISA and DAS-indirect ELISA. For DAS-ELISA the No. 1053 IgG
conjugate was used as second antibody. For DAS-indirect ELISA the unlabeled 3DF1
monoclonal antibody was the intermediate antibody followed by the goat anti-mouse IgG
conjugate. Samples were 200 /I of buffer extracts of greenhouse-grown Madam Vinous
sweet orange plants infected with the CTV isolates ground at a 1:10 dilution.


15
with mild isolates but unchallenged with T66a gave OD405 values
between 0.091 and 0.145 when analyzed with PCA. The
corresponding uninoculated healthy control plants averaged
0.039. The same treatments, including the healthy controls
gave values in the range of 0.011-0.025 when analyzed by DAS-
indirect ELISA with MCA-13. Treatments pre-inoculated with
mild isolates and further challenged with T66a gave OD405
values in the range of 0.130-0.217 with PCA and 0.145-0.189
with MCA-13. The control plants uninoculated with mild
isolates but challenged with T66a gave values of 0.174 with
PCA and 0.214 with MCA-13 (Table 2.1).
Also at warm temperatures, the Valencia/macrophylla
plants inoculated with the mild isolates and unchallenged with
T66a, gave OD405 values between 0.092 and 0.164 with PCA. The
value for the corresponding uninoculated healthy control
plants was 0.040. The same treatments, including the healthy
controls gave values in the range of 0.018-0.037 when analyzed
with MCA-13. Treatments pre-inoculated with mild isolates and
challenged with the T66a gave values in the range of 0.185-
0.311 with PCA and 0.104-0.305 with MCA-13. The control
plants uninoculated with mild isolates but challenged with
T66a gave values of 0.194 with PCA and 0.251 with MCA-13
(Table 2.2). The T30 isolate was not evaluated in this
portion of the experiment because not enough plants were
available. The plants pre-inoculated with the Tila isolate
were not protected and declined and died before the


95
Bar-Joseph, M., Marcus, R., and Lee, R.F. 1989. The continuous
challenge of citrus tristeza virus control. Ann. Rev.
Phytopathol. 27:292-316.
Bar-Joseph, M., Raccah, B., and Loebenstein, G. 1977.
Evaluation of the main variables that affect citrus
tristeza virus transmission by aphids. Proc. Intern. Soc.
Citriculture 3:958-961.
Bar-Joseph, M., Roistacher, C.N., and Garnsey, S.M. 1983.
Epidemiology and control of citrus tristeza virus. Pages
61-72, in: Plumb, R.T. and Thresh, J.M. (eds) Plant
Virus Epidemiology. Blackwell Sci. Pub. Oxford.
Bar-Joseph, M. Roistacher, C.N., Garnsey, S.M., and Gumpf,
D.J. 1981. A review on tristeza, an ongoing threat to
citriculture. Proc. Intern. Soc. Citriculture 1:419-423.
Beachy, R.N., Abel, P.P., Nelson, R.S., Register, J., Turner,
N.,and Fraley, R.T. 1987. Genetic engineering of plants
for protection against virus diseases. Pages 151-159, in:
Plant Resistance to Viruses. Ciba Foundation Symposium
133. Wiley, Chichester.
Bennett, C.W. and Costa, A.S. 1949. Tristeza disease of
citrus. J. Agr. Res. 78:207-237.
Ben-Ze'ev, I.S., Frank, A., and Bar-Joseph, M. 1988. Sensitive
detection of two plant viruses by enzyme-amplified ELISA.
Phytoparasitica 16:343-349.
Blue, R.L., Roistacher, C.N., G. Cartia, and Calavan, E.C.
1976. Leaf disc grafting: A rapid indexing method for
detection of some citrus viruses. Pages 207-212, in:
Calavan, E.C. (ed.). Proc. 7th Conf. Inter. Organ. Citrus
Virol. Riverside, California.
Brlansky, R.H. 1987. Inclusion bodies produced in Citrus spp.
by citrus tristeza virus. Phytophylactica 19: 211-213.
Brlansky, R.H. 1988. Other bacterial canker diseases. Page 7,
in: Whiteside, J.O., Garnsey, S.M., and Timmer, L.W.
(eds). Compendium of Citrus Diseases. APS Press. 80 p.
Brlansky, R.H., Garnsey, S.M., Lee, R.F., and Purcifull, D.E.
1984. Applications of citrus tristeza virus antisera for
use in labeled antibody, immuno-electron microscopical,
and sodium dodecyl sulphate-immuno-diffusion tests. Pages
337-342, in: Garnsey, S.M., Timmer, L.W., and Dodds, A.J.
(eds). Proc. 9th Conf. Intern. Organ. Citrus Virol.
Riverside, California.


43
(w/v) dilution. Microtiter plates were coated with 2.0 jug/ml
of purified CTV specific IgG in carbonate buffer (0.015 M
NaHC03, 0.03 M NaC03, pH 9.6) and incubated for 6 hr at 37C.
Antigen samples were added to the wells and incubated for 18
hr at 5C. CTV specific IgG conjugated to alkaline
phosphatase was used at a dilution of 1:1,000 in conjugate
buffer (PBS-Tween + 2% PVP + 0.2% bovine serum albumin) and
incubated for 4 hr at 37C (Bar-Joseph et al. 1979b, 1980).
The reaction with one mg/ml of p-nitrophenyl phosphate (Sigma)
in 10% triethanolamine, pH 9.8, was measured at 120 min at 405
nm (OD405) with a Bio-Tek EL-307 ELISA plate spectrophotometer.
Samples were considered positive when OD405 values were higher
than 0.100 or three times the mean of healthy controls,
whichever was greater. There were two replications per sample
in each microtiter plate. To estimate the relative CTV
concentration in test samples, a standard curve prepared by
diluting purified CTV T26 to OD260 values of 0.04, 0.02, 0.01,
0.005, 0.0025, 0.00125, and 0.0006 in a PBS-Tween + PVP
buffered extract of bark of healthy Citrus excelsa it was
included as a positive control in every test. Negative
controls included PBS-Tween + 2% PVP, conjugate buffer, and
extract from healthy C. excelsa. Madam Vinous sweet orange,
Mexican lime and grapefruit plants.
Results
Graft transmission of citrus tristeza virus isolates.
At 21 days post-inoculation the survival rate of grafted


54
A,
The reason why a low percentage of graft transmission of
the virus was found from some donor hosts, and the absence of
an expected 100% when the donor-receptor combination was of
the same species, is unknown. A possible explanation could
be differences in the virus distribution and/or concentration
in the donor tissues used as inoculum. Bark tissue contained
the highest antigen titer with OD405 values in the range of
0.221 and 0.349 in both C. excelsa and Madam Vinous with both
CTV isolates tested (Table 3.4). These values were, in some
instances, more than double those found in petioles and
midribs, and at least triple those found in the leaf blade.
Even though the statistical analysis did not show significant
differences in antigen titer in either different parts of the
same plant or from one plant to another, there were some
instances where OD405 values were as low as the healthy
controls. This indicates a possible absence of the virus in
those tissues and raises the possibility that occasionally
the tissue used for graft transmission may be virus-free, with
a subsequent failure in the transmission. Other possibilities
could be an occasional absence of phloem connections between
the donor and receptor tissues with a subsequent absence of
movement of the virus across the junction or the requirement
of a minimum of virus particles present in the tissue used as
inoculum in order to accomplish the transmission.
Citrus tristeza virus is phloem-limited (Bar-Joseph et
al. 1979a; Lister and Bar-Joseph, 1981), and is normally found


49
illustrated in Table 3.4. The average optical density values
at 405 nm (OD405) for bark tissue were 0.221 and 0.349 for the
T26 isolate and 0.266 and 0.336 for the T66a isolate in Madam
Vinous and C. excelsa. respectively. The OD405 values found in
the other tissues assayed in both hosts for T26 and T66a
isolates were in the range of 0.137 and 0.188 and 0.099 and
0.238 for petioles, 0.173 and 0.241 and 0.049 and 0.133 for
midribs, and 0.044 and 0.065 and 0.030 for leaf blades,
respectively. There were significant statistical differences
between C. excelsa and Madam Vinous for bark tissue with the
T26 isolate, and for both petioles and midribs with the T66a
isolate. The overall analysis showed that the highest OD405
values in both hosts for both T2 6 and T66a isolates, were
found in bark, followed by petioles and midribs. Leaf blades
showed the lowest OD405 values of all tissues assayed in both
hosts and isolates tested. Some differences in the OD405
values were found between different parts of the same plant,
and from one plant to another, in some virus isolate/host
combinations; however, the statistical analysis did not show
significative differences among them (data not shown). From
the standard curve prepared with purified T26 (Fig. 3.1), it
was estimated that an OD405 value of 0.4 65 was approximately
equivalent to 20 /g/ml of CTV, assuming an extinction
coefficient of 2.0 (Gonsalves et al. 1978). Therefore, the
CTV antigen concentration in the test samples (10 mg of


5
young trees propagated on sour orange has appeared in many
parts of Florida. Large scale outbreaks of induced decline
also have appeared in southern Florida, an area previously not
affected by CTV. Losses have exceeded 50% in some plantings
(Brlansky et al. 1986).
Management of CTV-induced decline in Florida is
difficult. The effective use of CTV tolerant rootstocks, such
as rough lemon (Citrus iambhiri Lush.), Troyer citrange
{Ppncirus trifoliata (L.) Raf. x C. sinensis (L.) Osb.},
Cleopatra mandarin (C. reshni Hort. ex Tanaka), sweet orange
(C. sinensis) and others (Grant et al. 1961; Wallace, 1978)
is diminished by their susceptibility to other diseases, most
importantly citrus blight, an endemic disease of unknown
etiology which is removing more than 500,000 trees from
production annually (Lee et al. 1988a).
The objectives of this research were: i) To evaluate some
CTV mild isolates under greenhouse conditions for cross
protecting ability against the decline syndrome, and ii) To
develop methods for detection of the severe CTV challenge
isolate in mixed infections.


the severe isolate in mixed infections. CTV-induced decline
(CTV-ID) occurred irregularly within the first 10 months after
challenge inoculation at both temperature regimens. The
preliminary evaluation of the cross-protecting ability of mild
isolates against the CTV-ID in plants on sour orange rootstock
can be accomplished under greenhouse conditions in a
relatively short time of 18-24 months.
Differences were found in the effectiveness of certain
tissues and/or hosts for graft-transmission of CTV. Leaf-
piece grafts transmitted CTV at a 90% rate vs a 75% rate using
bark pieces. Madam Vinous sweet orange was the most efficient
donor host giving 90% transmission to three receptor hosts,
followed by Mexican lime at 85%, and Citrus excelsa at 72%.
The dot-immunobinding assay (DIBA) was adapted for CTV
diagnosis by using several polyclonal and monoclonal
antibodies specific for CTV. The DIBA was as sensitive as
DAS-ELISA and DAS-indirect ELISA for CTV detection and
provides a reliable alternative for diagnosis of CTV.
xii


82
The 3DF1 MCA has been reported to react with a broad
spectrum of CTV isolates primarily on the basis of DAS-
indirect ELISA (Vela et al. 1986, 1988) In this work the
3DF1 reacted weakly with CTV isolates T26 and T66a in some
tests (Fig. 4.4, center, A2 and B3) and moderately in others
(Fig. 4.2, Cl,2,5,6). The reason for this is not known. For
isolate T26 it could be due to a low virus concentration as
was indicated by DAS-ELISA (Table 4.2). However, the T66a
isolate was at a high concentration as indicated in DAS-ELISA
with polyclonal antibody 1053 and in DAS-indirect ELISA with
MCA-13 (Table 4.2). Yet, in DAS-indirect ELISA 3DF1 gave a
low OD405 value for T66a (Table 4.2). This may indicate a
differential reactivity of the 3DF1 MCA with some particular
CTV isolates such as T66a.
The sensitivity limit for MCA-13 in both DIBA and DAS-
indirect ELISA was near 1/320 (Fig. 4.3 and Table 4.1). In
DIBA MCA-13 reacted strongly with CTV isolates T3, T36, T65a,
T66a, and T67a, all of which share severe biological
properties (Garnsey et al. 1987; Permar, et al. 1990; Rocha-
Pea and Lee, unpublished). However, there were slight
positive reaction with CTV isolates T50a and T55a which have
mild biological properties and with isolate T62a which has
severe properties (Rocha-Pea and Lee, unpublished) (Table
4.2) The MCA-13 has been reported to react specifically with
CTV isolates that have severe biological properties, such as
decline, stem pitting, seedling yellows, etc. (Permar and


60
and unfixed virus particles of T30, T3 6, and T2 6 CTV isolates,
respectively (R.F. Lee, unpublished) were used.
Immunoglobulins (IgG) were purified from whole sera by
the Protein A-Sepharose affinity chromatography method (Miller
and Stone, 1978) and adjusted to a final concentration of 1.0
mg/ml (OD280= 1.40) in PBS buffer with 0.02% sodium azide and
stored at 4C (Clark et al. 1986) The 3DF1 MCA was a gift
from Drs. P. Moreno and M. Cambra, Valencia, Spain, and its
preparation was described previously (Vela et al. 1986, 1988).
The MCA-13 that reacts specifically with severe CTV strains
(Permar et al. 1990) was a gift from Drs. T.A. Permar and S.M.
Garnsey. Goat anti-mouse and goat anti-rabbit IgG conjugated
with alkaline phosphatase were purchased from either
Boehringer or Promega.
The purified polyclonal IgG and 3DF1 MCA were tested at
concentrations of 1.0, 0.1, 0.2 or 0.01 /xg/ml. The MCA-13
was used as ascites fluid at a dilution of 1:5,000 (v/v) .
Goat anti-species IgG were used at concentrations recommended
by the manufacturer.
Sample preparation.- Bark tissue was peeled from fully-
expanded new flushes of CTV infected and healthy citrus
plants. The tissue was finely chopped and homogenized with
a Tekmar Tissumizer in the presence of Tris buffered saline
(TBS)-Tween (TBS = 0.02 M Tris, 0.5 M NaCl, pH 7.5), plus 0.5
% Tween 20 (TBS-Tween) at 1:10 (w/v) dilution.


84
strongest reactions, followed by TBS and carbonate. The
addition of Tween 20 gave slightly weaker reactions with all
buffers, but the spots were more uniformly spread on the
nitrocellulose. With both ELISA procedures the presence of
Tween 20 in the extraction buffers tended to increase the
OD405 values slightly. However, in DAS-indirect ELISA, the
presence of Tween 20 in the PBST more than doubled OD405
readings over PBS alone for CTV isolates T62a and T66a, but
had little effect with isolate T26. This did not occur with
TBS or the carbonate buffer. One possible explanation is that
the presence of Tween 20 in the PBST caused a conformational
change in the coat protein of T62a and T66a exposing more of
the epitope for binding with the 3DF1 MCA. This appeared to
occur to a much lower degree with isolate T26 in DAS-indirect
ELISA and with all three of these isolates reacting with the
polyclonal antibodies in DAS-ELISA. These results indicate
that PBS might be the best extraction buffer for DIBA, and
PBST or TBST might be the best for ELISA. However, it would
always be advisable to determine the effects of Tween 20 on
the serological methods and antigen/antibody combinations
under investigation.
There are several advantages to using DIBA over
conventional DAS-ELISA or DAS-indirect ELISA for CTV
detection. DIBA was rapid and easy to perform and, it was as
sensitive as either ELISA procedure for CTV diagnosis. The
entire test could be performed in 2-3 hours using polyclonal


24
T66a averaged a decline index of 4.0 (Table 2.5). The lowest
number of dead plants in the experiment (0/5) was obtained
when T26 was the protecting isolate (Table 2.5). At warm
temperatures, Valencia/macrophylla plants inoculated with mild
isolates, including the healthy uninoculated controls gave
decline indexes in the range of 1.7-2.2. In comparison, the
plants pre-inoculated with mild isolates and challenged with
T66a, showed higher decline index values in the range of 2.2-
9. The decline index scores for the control plants
uninoculated with mild isolates but challenged with T66a
averaged 6.8 (Table 2.6). The lowest number of dead plants
occurred when the T26 (0/5) and T55a (0/2) were the protecting
isolates (Table 2.6).
At cool temperatures, the decline index values for
Valencia/sour orange plants inoculated with mild isolates were
in the range of 1.0-3.2. The index of 3.0 for the healthy
uninoculated controls indicated the generally reduced growth
rate of plants at cool temperatures. The decline index values
for plants pre-inoculated with mild isolates and challenged
with T66a ranged from 5.0 to 7.5. The corresponding control
plants uninoculated with mild isolates but challenged with
T66a averaged a decline index of 7.5 (Table 2.7). The lowest
number of dead plants occurred when T26 (0/2) and T55a (1/4)
were the protecting isolates (Table 2.7).
Also at cool temperatures, the Valencia/macrophylla
plants inoculated with T26 and T55a isolates and the


17
Table 2.2 Relative antigen titer of citrus tristeza virus
(CTV) mild isolates in plants unchallenged and challenged by
the T66a severe isolate: II. Warm temperature,
Valencia/macrophylla.
Unchallenged17
Challenged
1/2/
CTV
OD*7
OD',05
isolate
Polyclonal
MCA-13
Polyclonal
MCA-13
Tila
0.092-7a-7
0.026 a
_6/
-
T26
0.132 a
0.030 a
0.185 a
0.104 a
T30
NE-7
NE
NE
NE
T55a
0.164 a
0.037 a
0.311 a
0.305 a
Healthy
0.040 a
0.018 a
0.194 a
0.251 a
or control
plants uninoculated
with mild isolate
17 One-year-old plants were graft inoculated with leaf
pieces under bark flaps on the stem from donor plants
infected with the indicated mild CTV isolates.
-7 After verifying virus infection with mild isolates by
DAS-ELISA with polyclonal antibodies, the challenged
plants were graft inoculated similarly with the T66a
severe isolate.
-7 Optical density at 405 nm (OD405) was measured after 120 min
of reaction.
Serological detection was carried out by DAS-ELISA with
polyclonal antisera and DAS-indirect ELISA with MCA-13
severe strain specific monoclonal antibody six months
after inoculation. Value is the average of duplicate
assays of the corresponding plants in Table 2.6.
-7 Numbers in the same column followed by different letters
are statistically different by Duncan's test (P < 0.05).
-1 = severely diseased plants died before serological
evaluation.
NE = treatment not evaluated.


103
citrus. Pages 58-62, in: Calavan, E.C. (ed). Proc. 7th
Conf. Intern. Organ. Citrus Virol. Riverside, California.
Yamaguchi, A. and Patpong, P. 1980. Comparison of time
requirement for graft transmission of citrus tristeza
virus with other fruit tree viruses. Pages 25-27, in:
Calavan, E.C., Garnsey, S.M., and Timmer, L.W. (eds).
Proc. 8th Conf. Intern. Organ. Citrus Virol. Riverside,
California.
Yokomi, R.K. and Garnsey, S.M. 1987. Transmission of citrus
tristeza virus by Aphis qossvpii and Aphis citricola in
Florida. Phytophylactica 19:169-172
Yokomi, R.K., Garnsey, S.M., Lee, R.F., and Cohen, M. 1987.
Use of insect vectors to screen for protecting effects
of mild citrus tristeza virus isolates in Florida.
Phytophylactica 19:183-185.
Yokomi, R.K., Garnsey, S.M., Permar, T.A., Lee, R.F., and
Youtsey, C.O. 1990. Natural spread of severe citrus
tristeza virus in citrus preinoculated with mild
isolates. In: Proc. 11th Conf. Intern. Organ. Citrus
Virol. Riverside, California (in press).


CHAPTER 1
INTRODUCTION
Citrus tristeza virus (CTV) is distributed in citrus
producing areas worldwide and is the most economically
important viral disease of citrus (Bar-Joseph et al. 1979a,
1981, 1989). The virus infects nearly all species, varieties,
and intergeneric hybrids of citrus, and some citrus relatives
(Bar-Joseph et al. 1979a; Garnsey and Lee, 1988; Muller and
Garnsey, 1984). However, the most destructive damage is the
induced decline in scions grafted on sour orange (Citrus
aurantium L.) rootstock. Some CTV isolates cause stem pitting
and loss of plant vigor on some orange and grapefruit scions
regardless of the rootstock (Bar-Joseph et al. 1979a, 1981,
1989) .
Citrus tristeza virus is a phloem-limited, flexuous
closterovirus approximately 2,000 x 11 nm in size, transmitted
by aphids in a semi-persistent manner (Bar-Joseph et al.
1979a; Lister and Bar-Joseph, 1981). A single stranded
positive sense RNA of 5.4-6.5 x 106 daltons has been isolated
from purified virus preparations (Bar-Joseph et al. 1985).
Several coat proteins of about Mr 28,000 (Guerri et al. 1990),
1