PURIFICATION, PARTIAL CHARACTERIZATION, AND SEROLOGY OF

THE CAPSID AND CYLINDRICAL INCLUSION PROTEINS OF

FOUR ISOLATES OF WATERMELON MOSAIC VIRUS

BY
ROBERT H. BAUM

A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF

THE UNIVERSITY OF FLORIDA

IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE

DEGREE OF DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA
1980

ACKNOWLEDGEMENTS

I wish to express my deepest appreciation to my wife, Diane, who
supported and championed me throughout this study and whose patience,
love, and understanding made light the burden of this experience. I
also wish to thank her for the excellent technical advice and expertise
which as an immunologist she was able to offer.

I wish to thank Dr. Dan Purcifull, chairman of my supervisor/
committee, for his support and counsel throughout this study and
particularly for his advice on the writing of this dissertation.

Appreciation is extended to other members of my supervisory
committee, Drs. Ernest Hiebert, John R. Edwardson, Daryl R. Pring,
Daniel A. Roberts, Robert E. Stall, and Indra K. Vasil for their help-
ful suggestions during the research and their constructive criticism
of the manuscript. Dr. Purcifull and Dr. Hiebert deserve special
thanks for their untiring efforts to instill in me a need for rigor
and objectivity in research. Dr. F. U. Zettler deserves an A-plus for
the excellent virology courses which he organized and for his ability
to convey to his students his own genuine excitement about virology. I
want to sincerely thank Mr. Richard G. Christie for providing an out-
standing example of what a scientist should be and also for his attempts
to teach this somewhat colorblind student how to recognize inclusions
in the light microscope. I also thank Richard Christie and Dr.
Edwardson for allowing me to cite some of their unpublished data on
Moroccan inclusion structure in this manuscript. The technical aid

ii

and optimism of Mr. W. E. Crawford is appreciated. I appreciate the
support of former graduate students in Plant Pathology and in particular
those in Plant Virology who never hesitated to pass on to others
knowledge which they had acquired. These are Dr. Albersio Lima, Dr.
Francisco Morales, Dr. David Thornbury, and Ms. Diana Zurawski. In
this same vein, I thank Dr. Mary Conde and Dr. Prem Chourey fox discussions,
advice, and friendship. Finally, I am thankful for again being able to
use the excellent typing skills of Ms. Donna Gillis in the final
preparation of this manuscript.

IX

TABLE OF CONTENTS

ACKNOWLEDGEMENTS ±±

LIST OF TABLES vi

LIST OF FIGURES vii

ABSTRACT

INTRODUCTION 1

LITERATURE REVIEW 4

WMV-1 and WMV-2 Defined 4

Transmission and General Characteristics 5

Differential Systemic and Local Lesion Hosts for

WMV-1 and WMV-2 3

Purification of WMV Isolates 10

Inclusions of WMV 13

Serological Relationships of WMV Isolates 16

MATERIALS AND METHODS . 20

Source of Virus Isolates 20

Maintenance and Propagation of Virus Isolates 20

Virus and Inclusion Purification 21

Purification of Cylindrical Inclusions for Peptide Mapping. . 25

Cleavage of WMV Capsid and Cylindrical Inclusion Proteins . . 27
Polyacrylamide Gel Electrophoresis of Viral and

Inclusion Proteins 28

Gel Electrophoresis of Cyanogen Bromide Cleaved Viral

Capsid Proteins 28

Serology 30

Cross-Absorption of Antisera 31

Serological Tests 32

Fractionation of Gamma Globulin for ELISA 33

Conjugation of Alkaline Phosphatase with

Gamma Globulin 34

Preparation of ELISA Plates 35

RESULTS 39

Purification and Properties of Watermelon Mosaic

Virus and Inclusions 39

Infectivity of Purified Viruses 48

Particle Length Determination of WMV-M 51

Molecular Weight Determination 51

Capsid Protein Digests by Cyanogen Bromide 58

Cylindrical Inclusion Digests 58

Serology 58

Enzyme-Linked Immunosorbent Assay (ELISA) 75

DISCUSSION 86

LITERATURE CITED 89

BIOGRAPHICAL SKETCH 96

LIST OF TABLES
Table

1 Geographical distribution of watermelon mosaic viruses. . . 6

2 Aphid transmissibility of watermelon mosaic viruses .... 7

3 Serological reactions with WMV virus antisera 63

4 Serological reactions with WMV inclusion antisera 65

5 ELISA serology of WMV isolates 84

LIST OF FIGURES

Figure

1 Scheme for determining optimum concentration of coating
y-globulin and enzyme labelled y-globulin 38

2 Flow diagram outlining the procedure for purification of

WMV-2 using n-butanol 41

3 Flow diagram outlining purification procedure of WMV
isolates and their cylindrical inclusions using

chloroform and carbon tetrachloride 43

4 Second stage in purification of the cylindrical inclusions

of WMV 45

5 Absorption spectra of purified preparations of WMV

isolates and WMV cytoplasmic inclusions 47

6 Electrophoretic analysis of purified undegraded WMV
capsid and cylindrical inclusion protein subunits in

an 8% polyacrylamide gel 50

7 Histogram of WMV-M particle lengths from a purified
preparation to show particle length from 500 to 900 nm. . . 53

8 Electrophoretic analysis of an 8% polyacrylamide gel
of purified WMV isolates stored at 4 C for three

weeks or longer 55

9 Electrophoretic analysis of an 8% polyacrylamide gel of
purified WMV isolates stored at 4 C for three weeks or

longer 57

10 Analysis of cyanogen bromide cleaved WMV capsid proteins

in a 12.5% polyacrylamide gel 60

11 Electrophoretic analysis of WMV cylindrical inclusion
proteins partially digested with Staphylococcus aureus

V-8 protease 62

12 Reciprocal SDS-double immunodiffusion tests between
WMV-1, WMV-2, and WMV-M with antisera obtained during
the first four months after initial injection of

immunogen 68

13 Double immunodiffusion serology of three WMV-M antisera
showing heterologous reactions with WMV-1 isolates 70

14 Intragel cross-absorption tests with WMV-1 Jordan and

WMV-1 Florida using SDS double immunodiffusion tests. ... 72

15 Reciprocal SDS double immunodiffusion tests between WMV-2
and 3CMV, B1CMV, and SoyMV and heterologous reactions

between WMV-2 antiserum and DMV and PVY 74

16 Reciprocal SDS double immunodiffusion tests of WMV
cylindrical inclusions 77

17 Heterologous reactivity of WMV-M inclusion antiserum with
other potyviruses by SDS double immunodiff fusion

serology 79

18 Intragel absorption of WMV-M inclusion antiserum with
inclusions of WMV-1, WMV-2, and WMV-M in an SDS
immunodiffusion medium 81

19 Intragel absorption of WMV-1 inclusion antiserum with
WMV-1 Jordan inclusions and WMV-1 Florida inclusions

in an SDS double immunodiffusion medium 83

Abstract of Dissertation Presented to the Graduate Council

of the University of Florida in Partial Fulfillment of the Requirements

for the Degree of Doctor of Philosophy

PURIFICATION, PARTIAL CHARACTERIZATION, AND SEROLOGY OF

THE CAPSID AND CYLINDRICAL INCLUSION PROTEINS OF

FOUR ISOLATES OF WATERMELON MOSAIC VIRUS

By

Robert H. Baum

March, 1980

Chairman: Dan E. Purcifull

Major Department: Plant Pathology

Watermelon mosaic viruses (WMV) , which are members of the potyvirus
group, have been classified into two distinct types, WMV-1 and WMV-2, on
the basis of host range, serology of the capsid proteins and morphology
of virus-induced inclusions. Several isolates, however, are at variance
with the two general types on the basis of host range or serology of
the capsid proteins. The purpose of this study was to further clarify
the distinctiveness of both WMV-1 and WMV-2 and an isolate from Morocco
(WMV-M) by the use of serology and peptide maps of the capsid and
cylindrical inclusion proteins.

Virus isolates used extensively in this study were WMV-M, WMV-1
and WMV-2 from Florida, and WMV-1 from Jordan. Viruses were increased
in pumpkin, Cuauvbita vevo L. "Small Sugar," and purified by clarification
of sap with a mixture of chloroform and carbon tetrachloride followed by
concentration with polyethylene glycol and fractionation on either
cesium chloride or cesium sulfate isopycnic density gradients. Cyto-
plasmic cylindrical inclusions were purified using the same clarification
procedure followed by fractionation on sucrose step gradients or
preparative polyacrylamide gels.

ix

Antisera to both formaldehyde fixed and unfixed virus and to
purified inclusions were produced in New Zealand white rabbits. Sera
collected following immunization up to approximately one year were
checked for immunochemical specificity.

The serological relationships of selected potyviruses were
determined by enzyme linked immunosorbent assay and sodium dodecyl
sulfate (SDS) immunodiffusion. In reciprocal SDS immunodiffusion tests,
heterologous reactions were obtained between WMV-2 and: bean common
mosaic virus (BCMV) , blackeye cowpea mosaic virus (B1CMV) , and potato
virus Y (PVY) , but not WMV-1 (Jordan or Florida) , WMV-M, or papaya
ringspot virus (PRSV) . WMV-1 Florida and WMV-1 Jordan, hereafter
collectively referred to as WMV-1, were shown to be serologically
identical by intragel absorption in SDS immunodiffusion tests. WMV-1
reacted heterologously with PRSV infected sap but not in reciprocal
immunodiffusion tests with WMV-2, WMV-M, BCMV, B1CMV, or PVY. Sera
from early bleedings of WMV-M did not react with WMV-1 or WMV-2, but
bleedings after four months did give reactions of partial identity with
WMV-1. This was confirmed by intragel absorption and ELISA.

Antisera specific for the cylindrical inclusions of WMV-M, WMV-1,
and WMV-2 in SDS immunodiffusion cests with sap and purified inclusions
were obtained. Antisera to WMV-2 inclusions reacted heterologously
with BCMV, dasheen mosaic virus (DMV) , soybean mosaic virus (SoyMV) ,
and lettuce mosaic virus (LMV) . WMV-M inclusion antiserum reacted
heterologously with WMV-1, WMV-2, DMV, and PRSV and with some isolates
of LMV. The WMV-1 inclusion antiserum did not react with either WMV-M
or WMV-2 inclusions.

Analysis of peptide fragments after cleavage of capsid proteins
with cyanogen bromide (CNBr) showed distinct patterns for WMV-1, WMV-2,
and WMV-M. Staphylococcus aureus V-8 protease digests of the cylindrical
inclusions of WMV-1, WMV-2, and WMV-M were analyzed by disc poly-
acrylamide gel electrophoresis. While there were some similarities
between patterns, significant differences existed.

This study provides additional evidence that WMV-1 and WMV-2 are
distinct viruses on the basis of inclusion serology and analysis of the
peptides formed by partial cleavage of capsid and cylindrical inclusion
proteins. Evidence is also provided that WMV-M is a distinct virus.

INTRODUCTION

Watermelon mosaic viruses, which are members of the potyvirus
group (Brandes, 1964; Harrison et al. , 1971), cause economically
important diseases of cucurbits throughout the world (van Regenmortel,
1971) . The watermelon mosaic viruses have f lexuous anisometric particles
700-800 nm long, induce cylindrical inclusions in their hosts, and are
transmitted in a nonpersistent manner by aphids (van Regenmortel, 1971;
Edwardson, 1974a).

Symptoms induced by WMV are highly variable, depending on the
virus strain, the host, and on environmental conditions. Symptoms range
from mild mottling to severe shoestring distortion and blistered ;leaves
(Anderson, 1954; Webb and Scott, 1965; Greber, 1969; Milne et al. , 1969;
Bakker, 1971; Thomas, 1971a; van Regenmortel, 1971; Webb, 1971; Demski
and Chalkley, 1974; Fischer and Lockhart, 1974; Horvath et al., 1975).
Due to variability, this characteristic cannot be readily used to
distinguish watermelon mosaic from other viral diseases of cucurbits.
Serological tests have been used to distinguish six potyviruses that
infect cucurbits (Purcifull and Hiebert, 1979). Other viruses that
infect cucurbits are distinguished from WMV on the basis of the host
range (Lindberg et al., 1956; Grogan et al., 1959), mode of transmission
(Greber, 1969), absence of cylindrical inclusions in the cytoplasm of
infected plants (Christie and Edwardson, 1977), physical properties
(Lindberg et al. , 1956), serology, and virus morphology (van Regenmortel
et al. , 1962).

1

Watermelon mosaic virus was first described by Anderson (1954)
and it was first purified by van Regenmortel (1961) . The North American
isolates of the virus were divided into watermelon mosaic virus 1 (WMV-1)
and watermelon mosaic virus 2 (WMV-2) on the basis of the host range
(Webb and Scott, 1965). WMV-1 isolates were limited to the Cucurbitaceae,
whereas WMV-2 had a wider host range covering some 17 plant families
(van Regenmortel, 1971) . WMV-1 and WMV-2 were reported to be serologi-
cally distinct (Webb and Scott, 1965) and reciprocal tests indicated
that neither could cross-protect against a challenge by the other virus
(Webb and Scott, 1965). It was concluded, therefore, that WMV-1 and
WMV-2 are distinct viruses. The serological results have been con-
firmed (Purcifull and Hiebert, 1979; Greber, 1978; Baum et al., 1979).
However, Milne and Grogan (1969) reported a close serological relation-
ship between WMV-1 and WMV-2. On the basis of serology, cross-protection,
indicator hosts, and host range, Milne and Grogan concluded that their
WMV-1 and WMV-2 isolates were strains of the same virus. Webb and
Scott (1965) suggested the possibility of a third virus or virus group
because they were unable to get a positive serological reaction between
WMV-1 or WMV-2 and van Regenmortel' s South African WMV antisera. The
presence of a third member of the watermelon mosaic virus subgroup
was also suggested by Schmelzer (1969) and by Horvath et al. (1975)
based on differences in host range and symptoms and by Purcifull and
Hiebert (1979) based on serological experiments. They were
unable to detect a serological relationship between a
Moroccan isolate of WMV (WMV-M) (Fischer and Lockhart, 1974)
and antisera to Florida isolates of WMV-1 and WMV-2.

Although antisera to the cylindrical inclusions have been useful
in studying relationships of other potyviruses (Hiebert et ai., 1971;
Purcifull et ai., 1973; McDonald and Hiebert, 1975), the serological
relationships of WMV- induced inclusions have not been studied previously.

The objectives of this dissertation were to: (i) study the sero-
logical relationship of WMV-M to Florida isolates of WMV-1 and WMV-2 and
to a Jordan isolate of WMV-1, (ii) test the use of WMV cylindrical
inclusion antisera in determining relationships among isolates of WMV,
and (iii) analyze the relationships among WMV isolates by comparing
peptide maps of capsid and cylindrical inclusion proteins.

LITERATURE REVIEW

Watermelon mosaic virus was described by Anderson (1951a, 1951b,
1954), who recognized that it was distinct from other cucurbit viruses.
WMV was first noted to be distinct from cucumber mosaic virus (CMV) in
spite of what appeared to be partial cross-protection (Anderson, 1951a).
These distinguishing factors were: a synergistic effect resulting in a
more severe disease when both viruses were present in cucumber
(Cuoumis sativus L.), increased numbers of primary lesions induced by
CMV on watermelon (Citrullus lanatus [Thunb.] Matsun. and Naki) when
inoculation and infection occurred with both viruses, and no evidence
that CMV predominated over WMV (Anderson, 1951a).

The symptoms observed by Anderson on sy s t enica'ily infected leaves
of cucurbits were mild chlorosis, mottle, green vein banding, raised
blisters, distortion, and shoestringing. These have generally been
the symptoms observed on WMV infected cucurbits, but fruit distortion,
stunting, and mottling are often noted — especially on plants infected
early in the season (Thomas, 1971b; Demski and Chalkley, 1974; Fischer
and Lockhart, 1974).

WMV-1 and WMV- 2 Defined

Webb and Scott (1965) divided WMV into two distinct viruses,
partly on the host range of North American isolates. WMV-1, as defined
by Webb and Scott (1965) infects only the Cucurbitaceae while WMV-2
can infect plants in 21 additional families (Edwardson, 1974b; Molnar
and Schmelzer, 1964; Webb and Scott, 1965; Greber, 1969; Prowidenti

4

and Schroeder, 1970; Webb, 1971). WMV-2 usually produces milder leaf
symptoms with less distortion and blistering than WMV-1 (Webb and
Scott, 1965). Both viruses have a worldwide distribution (Table 1),
though WMV-1 may have been recently introduced into Europe (Horvath
et al., 1975) and Australia (Greber, 1978) from the western hemisphere.

Transmission and General Characteristics

Seed transmission of WMV was not demonstrated in a test using
several hundred seeds from infected Summer Crookneck Squash (Cuaurbita
pepo. var. melopevo Alef.) and cantaloupe (Cucumis melo L.) (Anderson,
1951b). Seed transmission of WMV has not been detected by others using
different isolates of WMV and other cucurbit hosts (Anderson, 1951b;
Grogan et al. , 1959; Greber, 1969; Thomas, 1971a; Fischer and Lockhart,
1974; Bhargava, 1977). WMV may therefore be distinguished from viruses
which are seed-borne in cucurbits. These include squash mosaic virus
(SqMV) (Campbell, 1971), and melon necrotic spot virus (MNSV)
(Gonzalez-Garza et al. , 1978).

WMV-1 and WMV-2 have been transmitted in a nonpersistent manner
by several aphid species (Table 2). Aphis gossypii Glover transmits
WMV, single aphids being capable of transmitting WMV to about 50% of
test plants after an acquisition period of 18-36 seconds (Anderson,
1951b). Thomas (1971a) was unable to transmit WMV-2 from infected to
healthy plants of Buttercup Squash {Cuauvbita maxima Duch.) using A.
gossypii or Myzus pevsiaae (Sulzer) . Bakker (1971) also reported the
nontransmissibility of a Kenyan isolate of WMV by Aphis fabae.

The physical properties observed by Anderson (1954) have been
confirmed for both WMV-1 and WMV-2. The physical properties are: a

Table 1.

Geographical distribution of watermelon mosaic
viruses.

Australia

WMV-1

WMV-2

Canada

WMV-2

Chile

WMV-2

Cuba

WMV-1

Czechoslovakia

WMV-2

Eastern Europe

WMV-1

WMV-2

WMV-?

France

WMV-2

Germany

WMV-1

Guadeloupe

WMV-1

Hungary-

WMV-2

India

WMV-1

WMV-1

WMV-2

Iran

WMV-1

WMV-2

Iraq

WMV-2

Israel

WMV-2

Italy

WMV-1

WMV-2

Japan

WMV-2

Kenya

WMV- Kenya

Mexico

WMV-1

WMV-2

Morocco

WMV-Morocco

New Zealand

WMV-2

South Africa

WMV-SA*

United States

Arizona

WMV-2

California

WMV-1

WMV-2

Florida

WMV-1

WMV-2

Georgia

WMV-2

Hawaii

WMV-1

WMV-2

WMV-Kauai**

Massachusetts

WMV-2

New York

WMV-1

WMV-2

U.S.S.R.

WMV-2

Venezuela

WMV-1

WMV-2

Yugoslavia

WMV-2

Greber, 1978

Greber, 1978

Gates and Bronskill, 1976

Auger et at. , 1974

Schmelzer, 1969

Schmelzer and Milicic, 1966

Molnar and Schmelzer, 1964

Molnar and Schmelzer, 1964

Molnar and Schmelzer, 1964

Arteaga et at. , 1976

Hein, 1977

Quiot et at. , 1971

Horvath et at. , 1975

Bhargava, 1977

Ghosh and Mukhopadhyay , 1979

Ghosh and Mukhopadhyay, 1979

Ebrahim-Nesbat , 1974

Ebrahim-Nesbat, 1974

Shawkat and Fegla, 1979

Cohen and Nitzany, 1963

Ragozzino and Stefanis, 1977

Ragozzino and Stefanis, 1977

Inouye, 1964

Bakker, 1971

Milne and Grogan, 1969

Milne and Grogan, 1969

Fischer and Lockhart, 1974

Thomas, 1971a

van Regenmortel, 1961

Nelson and Tuttle, 1969
Milne and Grogan, 1969
Milne and Grogan, 1969
Adlerz, 1969
Adlerz, 1969
Demski, 1968
Shanmugasundaram et at. .

Shanmugasundaram et at

1969
1969
1969

Shanmugasundaram et at,
Komm and Agrios , 19 78
Prowidenti and Schroeder, 1970
Prowidenti and Schroeder, 1970
Schmelzer and Milicic, 1966
Lastra, 1968
Lastra, 1968
Stakic and Nikolic, 1966

.WMV-SA = South African isolate
WMV-Kauai isolates from Kauai were distinct from WMV-1 and WMV-2.

C 0) r- .Q ^H

co a\ H co J

5.

cd

00
M

—

•t

e-i

—1

«

.« +i

> oc

m

CI

G

«

o> <a

CO 0)

u

ca

»-i

vo

SO Pn

u

-J

4J

o-. jr

h

r,

'.»

4J

CO

rH 4J

Cd T>

a

id

03

cd

A C

J

4J

^

h-l

« >

pa cd

0

*

C s-i

■H

cd

cd o

•« LJ

in

3

,X3

in

00 33

so cd

^D

o*

Cfl

^o

o .«

r- M

31

o>

!-i sr

ON 3

K

vO

r»

W

cr\

o

~l

H

0

U

c/3.

(3

c

cd

■H

T3

so

0)

H O r*> H cd

M C «^ (?

-G "0 !**• iJQ ^
0) C O (U ON

2 CU H gH

jcs a n j

Oj -H C

>- cd a) -C

< O

M

M

CIJ

^

1J

u

r».

^

IS

M

cn

r-.

Jrf

cd

CB

cd

CT\

cd

^

►J

p— 1

T-t

pq

CZ)

o> gn o> on

En U PQ O O

<"

a.

-3 -5
dilution end-point of approximately 10 to 10 , a thermal inactivation

point of about 55 C to 65 C, and a longevity in vitro of about 6 to 20

days (van Regenmortel ei at., 1962; Webb and Scott, 1965; Milne and Grogan,

1969; Greber, 1978).

Differential Systemic and Local Lesion Hosts for
WMV-1 and WMV-2

Pinpoint, brown-bordered, local, circular, paper-white necrotic
lesions with minute dark brown centers developed on cotyledons and leaves
of muskmelon (Cucumis melo L. var. retiaulatus Naud.) selection P.I.
180280) inoculated 4-6 days earlier with two isolates of WMV (Webb,
1963). Some plants that developed these local lesions remained free of
virus in secondary leaves whereas others died after developing stem and
top necrosis. Muskmelon cotyledons inoculated with other isolates of WMV
developed a systemic mottle but not local lesions. In reciprocal tests,
no cross-protection occurred with local-lesion and systemic mottle
isolates (Webb, 1963), suggesting that WMV might consist of at least
two unrelated viruses. Further study showed that isolates that induced
local lesions on muskmelon were also restricted to the Cucurbitaceae;
these isolates were hereafter classified as belonging to the WMV-1
group. Those isolates which induced a systemic mottle in muskmelon
also had wider host ranges, including certain species in the families
Leguminosae, Ghenopodiaceae, and Euphorbiaceae, and were designated
members of the WMV-2 group (Webb and Scott, 1965). The WMV-2 isolates
were maintained free of WMV-1 by culturing the former in non-cucurbitaceous
hosts (Webb, 1965).

Toba (1962) used Chenopodium amavantiaolor Coste and Reyn. in a
host range study of WMV isolates in Hawaii. All of his isolates were

restricted to the Cucurbitaceae and did not form local lesions on C.
amc&antiooZov. Cohen and Nitzany (1963) studied cucurbit viruses in
Israel. Based on its ability to induce lesions in C. amarantiaolov ,
to infect legumes, and on its physical properties, the melon mosaic
virus they reported was of the WMV-2 type. Molnar and Schmelzer (1964)
in an extensive study of the host range of two Eastern European isolates
of WMV, found that several members of the Chenopodiaceae formed local
lesions in response to inoculation with one of these strains but not
with the other. Both strains, however, infected many families outside
the Cucurbitaceae, indicating that not all WMV-2 isolates will form
local lesions on C. amaranticolor. This inability of some isolates of
WMV to induce local lesions on C. amavantiaolov led to the identification
of a virus which was latent in cucurbits and named cucurbit latent virus
(CLV) by Webb and Bohn (1961). CLV formed local lesions in C.
amavantioouor and had a host range similar to WMV-2. Lack of cross-
protection in tests involving CLV infections challenged by several other
viruses (SqMV) , CMV, tobacco ringspot virus [TRSV] and WMV) indicated
that CLV was unrelated to these other viruses. Separation of WMV
into WMV-1 and WMV-2 was not recognized in 1961 and Webb and Bohn did
not indicate what isolate of WMV was used in the cross-protection tests.
That CLV was probably synonymous with WMV-2 was suggested by studies
in which it was impossible to separate or differentiate WMV-2 from
CLV (Milbrath and Nelson, 1968; Demski, 1968; Milne et at., 1969). All
of Milne and Grogan's WMV-2 isolates formed local lesions on C.
amaranticolor' and both WMV-2 and CLV were systemic in the malvaceous
plant Lavatsra tvimestri-s L. (Milne et at. , 1969) .

10

Separation of WMV-1 from WMV-2 was achieved by using Luffa
acutangula Roxb. , which was found to be susceptible to WMV-1, but immune
to the WMV-2 isolates tested (Webb, 1965). Symptoms induced by
inoculation of cotyledons with WMV-1 were expressed as mild to severe
chlorotic spots with stunting of leaves and runners. This use of L.
acutangula to separate WMV-1 from WMV-2 has been questioned by Milne
et at. (1969), who found it to be susceptible to 16 of 46 isolates
of WMV-2 tested. Several other workers have since used Luffa to
separate WMV-1 from WMV-2 without difficulty (Greber, 1969; Quiot
et at., 1971; Arteaga et at., 1976; Purcifull and Hiebert, 1979).
Bhargava (1977) found that certain cultivars of Luffa aautanguZa
were susceptible to two different isolates of WMV-2. These two
isolates, however, were primarily limited to the Cucurbitaceae. Many
species outside the Cucurbitaceae were tested by Bhargava, but only
Vigna sinensis Savi ex Hasski var. Black Turtle was susceptible to
both isolates and Zinnia elegans Jacq. was a symptomless carrier of one
of the isolates. By the use of reciprocal tests, Webb (1971) was unable
to achieve infection with 17 isolates of WMV-2 in L. aautangula by
approach grafting to watermelon or cantaloupe.

Purification of WMV Isolates

Several schemes have been developed or modified for purifying
WMV. van Regenmortel (1961) developed a purification procedure for
several South African isolates of WMV, testing different methods for
clarifying and concentrating infective plant sap. Freezing or treat-
ment of sap with ethanol greatly reduced infectivity and antigenicity.
The use of n-butanol or a mixture of c .loroform and n-butanol was less

11

effective than chloroform alone for clarification of sap in phosphate
buffer. Three cycles of differential centrifugation were superior in
removing host material as compared with perosmosis, pervaporation,
salting out with ammonium sulfate or acid precipitation with 10% acetic
acid. Following the removal of most host material, the virus was
further purified by zone electrophoresis (van Regenmortel, 1960). Even
though this method was superior to DEAE chromatography, aggregation
sometimes occurred. Preparations of WMV obtained by this method were
satisfactory for the production of antisera but were too aggregated for
determination of particle lengths. An alternative method was developed
(van Regenmortel et at. , 1962) , in which the leaf sap was clarified
with chloroform in sodium citrate buffer followed by three cycles of
differential centrifugation. The concentrated, clarified virus suspen-
sion was filtered through a column consisting of 4% granulated agar in
sodium citrate buffer. This method removed all visible impurities and
resulted in a nonaggregated virus preparation suitable for particle
length determination, as well as for the production of antisera.

Webb and Scott (1965) clarified infective sap in sodium phosphate
buffer by centrifugation at 5000 g followed by two cycles of differential
centrifugation. The resuspended virus had been concentrated thirty
times based on an infectivity dilution end point of 10 after purifica-
tion of WMV-1. They were successful in purifying WMV-1 and producing
an antiserum against it, but were unable to purify sufficient virus for
the production of an antiserum against WMV-2.

Milne and Grogan (1969) used potassium phosphate buffer to
homogenize pumpkin (Cuaurbita pepo L. 'Small Sugar') leaves infected

12

with either WMV-1 or WMV-2. The resulting sap was clarified using
n-butanol followed by two cycles of differential centrifugation. The
virus was further purified by zone electrophoresis (van Regenmortel,
1960). Neither the WMV-1 nor WMV-2 preparations reacted with an anti-
serum made against concentrated healthy pumpkin sap. Antisera were
made against the purified WMV-1 and WMV-2. The virus was severely
aggregated after zone electrophoresis, and leaf dips (Hitchborn and
Hills, 1965) were used in normal length determinations of virus
particles.

Purcifull and Hiebert (1979) clarified WMV-1 sap from infected
Small Sugar Pumpkin with a chloroform-carbon tetrachloride mixture in
phosphate buffer. Butanol was used as the organic solvent for clarifying
sap from WMV-2 infected pumpkin leaves. The virus was precipitated
from the aqueous phase with polyethylene glycol and subjected to
equilibrium density gradient centrifugation in CsCl (p = 1.28 g/ml).
The virus-containing zone was diluted with buffer and subjected to one
cycle of differential centrifugation. These purified virus preparations
were used for determinations of the molecular weights of the capsid
proteins and for the production of antisera. The molecular weights of
undegraded capsid proteins of WMV-1 and WMV-2 were about 36,500 daltons.

Several other workers have used simplified or shortened pro-
cedures to purify WMV-2. Thomas (1971a) and Auger et at. (197^) used
modifications of Milne and Grogan's purification scheme using n-butanol
clarification to purify WMV. Bakker (1971) and Bhargava (1977) used
chloroform clarification to purify isolates of WMV. Fischer and
Lockhart (1974) modified the procedure of Damirdagh and Shepherd

13

(1970) to partially purify a Moroccan isolate of WMV. Thomas (1971a)
used the purified WMV-2 to produce an antiserum. None of these workers
used zone electrophoresis or cesium chloride equilibrium density gradient
centrifugation in their purification schemes.

Inclusions of WMV

Potyviruses induce the formation of distinctive cylindrical
inclusions in the cytoplasm of infected host cells (Edwardson, 1966;
Purcifull and Edwardson, 1967; Hiebert et at., 1971; Edwardson, 1974a).
Cylindrical inclusions are composed of protein (Shepard, 1968) whose
monomeric subunits have estimated molecular weights of approximately
67,000 to 70,000 daltons (Hiebert and McDonald, 197 3) and are immuno-
logically distinct from the viral capsid protein (Shepard and Shalla,
1969; Purcifull et at., 1973; McDonald and Hiebert, 1975) and host
proteins (Purcifull et at. , 1973). Edwardson (1966) proposed that the
presence of cytoplasmic cylindrical inclusions was diagnostic for
infection by members of the potyvirus group. Differences in morphology
of the inclusions as determined by ultrastructural studies (Edwardson
et at. , 1968) led to the separation of potyviruses into three sub-
divisions (Edwardson, 1974a). Viruses in Subdivision I, to which WMV-1
belongs, induce tubular inclusions attached to the central portion of
the cylindrical inclusion. In cross section, the inclusions appear
as scrolls, while in longitudinal section they appear as tubes. Sub-
division II viruses form laminated aggregate inclusions attached to
the central portion of the cylindrical inclusion. The laminated
a§gregates are usually observed in negatively stained preparations
as roughly triangular or rectangular plates appressed together for part

14

or all of their length. Viruses in Subdivision III, to which WMV-2
belongs, induce both tubes and laminated aggregates in their host
cells.

Martelli and Russo (1976) found several isolates of WMV which
infected C. amarantiaolov and C. quinoa. They therefore classified
these isolates as WMV-2 since their host ranges extended outside the
Cucurbitaceae. These isolates, however, induced tubular inclusions,
(as reported for Subdivision I types) and amorphous cytoplasmic in-
clusions, as reported for WMV-1 (Edwardson, 1974; Christie and Edward-
son, 1977). Purcifull and Hiebert (1979) tested two of Martelli and
Russo 's isolates and found that they were serologically identical to
WMV-1 but not to WMV-2, gave a systemic reaction when inoculated to
Luff a aautangula, and did not infect either Niaotiana benthamiana
(Christie and Crawford, 1978) or Pisim sativum. Hence the question
of the reliability of using the Chenopodiaceae for typing WMV isolates
is probably more relevant than considering Martelli and Russo 's
isolates as exceptions to Edwardson 's scheme for separating potyviruses
on the basis of inclusion morphology.

Other types of inclusions are associated with certain potyviruses
(Edwardson, 1974a; Christie and Edwardson, 1977) . In addition to
tubular inclusions, WMV-1 induces amorphous inclusions similar to those
induced by papaya ringspot virus (Edwardson, 1974a). A thin plate-like
nuclear inclusion has been detected in cells infected with WMV-2, by
both light and electron microscopy (Christie and Edwardson, 1977) .
These nuclear inclusions have not been observed in plants infected
with WMV-1 isolates (Edwardson, 1974a; Christie and Edwardson, 1977).

15

Many virus induced inclusions can be observed and identified with
the light microscope after relatively simple staining techniques (Christie,
1967; Christie and Edwardson, 1977). These procedures allow for the
rapid screening of infected tissue at minimal time and expense. Often,
accurate identification to potyvirus subgroup or even to a specific
virus is possible. The cytoplasmic and nuclear inclusions induced by
WMV isolates can be detected by these techniques (Christie and Edward-
son, 1977).

The serological properties and relationships of some potyvirus
cylindrical inclusions have been investigated (Hiebert et at. , 1971;
Purcifull et at., 1973; Batchelor, 1974; McDonald and Hiebert, 1975;
Purcifull and Batchelor, 1977; Lima, 1978; Zurawski, 1979). Antisera
produced against partially purified inclusions of tobacco etch virus
(TEV) and potato virus Y (PVY) gave strong homologous reactions but did
not cross react with each other (Hiebert et at., 1971).

Five potyviruses (TEV, PVY, turnip mosaic virus, bidens mottle
virus, and pepper mottle virus) were found to induce cylindical inclusions
that were serologically distinct (although some were related) and the
propagative hosts did not affect the antigenic specificity of the
inclusions (Purcifull et at., 1973). Antigenic differences between
strains of turnip mosaic virus (TuMV) (McDonald and Hiebert, 1975) were
detected in the capsid proteins but not the cylindrical inclusions even
though one of the three strains studied had distinctly different
laminated aggregates. This work supported the concept that cylindrical
inclusions are coded for by the viral nucleic acid and that serological
studies of inclusions could be useful in determining the taxonomic
relationship between potyviruses.

16

Direct evidence for the hypothesis that cylindrical and some
nuclear inclusions are products of the potyvirus genome come from in
vitro translation of pepper mottle virus (PeMV) and TEV RNAs (Dougherty,
1979). In these studies, molecular weight determinations utilizing
SDS polyacrylamide gel electrophoresis and serology indicate cylindrical
inclusion and capsid proteins are synthesized in vitro from PeMV and
TEV RNA.

Serological Relationships of WMV Isolates

There have been several points of controversy concerning serological
relationships among WMV isolates. Specifically, there have been dis-
agreements about the serological relationships between WMV-1 and WMV- 2
(Webb and Scott, 1965; Milne and Grogan, 1969; Purcifull and Hiebert.
1979) , and about the serological relationship of papaya ringspot virus
to WMV-1 and WMV- 2 (Milne and Grogan, 1969; Purcifull and Hiebert,
1979). There also have been indications that at least one serotype
distinct from either WMV-1 or WMV- 2 may exist (Webb and Scott, 1965;
Purcifull and Hiebert, 1979). Complicating interpretation of the
various results is the use of different virus isolates for the prepar-
ation and testing of antisera, and the use of various types of
serological tests.

Webb and Scott (1965) divided WMV into two groups (WMV-1 and
WMV- 2) , partly on the basis of serological differences between the two
types. Milne and Grogan (1969), however, reported that WMV-1 and
WMV-2 were serologically very closely related and they concluded that
WMV-1 and WMV-2 should be considered as strains of the same virus.
Several subsequent workers have been unable to find a close serological

17

relationship between WMV-1 and WMV-2 (Bakker, 1971; Greber, 1978;
Purcifull and Hiebert, 1979). Purcifull and Hiebert (1979) produced
antisera to Florida isolates of WMV-1 and WMV-2. In SDS double
immunodiffusion tests, the WMV-1 antiserum gave a positive reaction
only with isolates of WMV-1. Likewise, the WMV-2 antiserum reacted
only with WMV-2 isolates. Martelli and Russo (1976) found several
isolates of WMV from the Mediterranean region which were limited in
host range to the Cucurbitaceae and Chenopodiaceae. Purcifull and
Hiebert (1979) tested two of these Mediterranean isolates and found
that they reacted with WMV-1 antiserum but not with WMV-2 antiserum.

In addition to serological differences between WMV-1 and WMV-2,
there has been conflicting evidence on the serological relationship
between WMV-1 or WMV-2 and several African isolates of WMV. Some of
this evidence has led to the suggestion that WMV may consist of three
or more serologically distinct viruses. Webb and Scott (1965) suggested
the presence of a third serologically distinct virus in the WMV group
when they were unable to obtain positive serological reactions in tests
with an antiserum to a South African isolate of WMV and antigens of
North American WMV-1 and WMV-2 isolates. Lastra (1968) obtained
positive reactions in microprecipitin tests between Venezuelan isolates
of WMV-1 and WMV-2 and antiserum obtained from Grogan and thought to have
been made against WMV-1. He was, however, unable to get a positive
reaction using the same isolates and antisera specific for the South
African isolates. Bakker (1971) found that a Kenyan isolate of WMV,
which was limited in host range to the Cucurbitaceae and Chenopodiaceae,
reacted with Milne's WMV-2 antiserum and van Regenmortel' s South African

18

WMV antiserum but not with Milne's WMV-1 antiserum. Milne and Grogan
(1969), however, state that their WMV-1 and WMV-2 isolates gave
positive reactions with the South African WMV antiserum. Purcifull
and Hiebert (1979) got negative reactions between WMV-M and antisera
specific for Florida isolates of WMV-1 or WMV-2. They were unable to
produce an antiserum to WMV-M and therefore considered the possibility
of WMV-M representing a third serotype as tentative. Schmelzer (1969)
suggested the possibility of a third type of WMV based on host range
differences and symptom expression of the South African isolates but
did not present serological evidence.

Serological relationships between WMV and other potyviruses have
been established though some confusion has resulted due to lack of
agreement as to whether WMV-1 and WMV-2 are serologically distinct.
Milne and Grogan (1969) found WMV-1 and WMV-2 to be closely related to
papaya ringspot virus, while Purcifull and Hiebert (1979) obtained
reactions of identity between papaya ringspot virus and WMV-1 antiserum
but did not get a reaction with WMV-2 antiserum. Purcifull and Hiebert
also showed in reciprocal immunodiffusion tests that soybean mosaic
virus was closely related to but distinct from WMV-2. Van Regenmortel
ei at. (1962) demonstrated a serological relationship using an antiserum
specific for the South African isolate of WMV and both BYMV and potato
virus Y (PVY) . Several other potyviruses which have been reported to
infect cucurbits, viz. LMV, TuMV, and the severe strain of BYMV, were
found to be serologically distinct from WMV-1 and WMV-2 (Florida
isolates) in reciprocal gel immunodiffusion tests performed by Purcifull
and Hiebert (1979). No cross-reactions were observed in reciprocal

19

tests between WMV-1 and soybean mosaic virus. Blackeye cowpea mosaic
virus (B1CMV) was shown in reciprocal immunodiffusion tests to be related
to but distinct from WMV-2. No cross-reactions were observed in
reciprocal immunodiffusion tests between B1CMV and WMV-1 (Lima et at. ,
1979).

Milne and Grogan (1969) obtained precipitin bands when detergent-
treated, partially purified papaya ringspot virus (PRSV) was tested with
either WMV-1 or WMV-2 antisera. They observed no detectable serological
difference between PRSV and WMV-1 or WMV-2 antigens treated in the same
manner. Purcifull and Hiebert (1979) did not get a detectable reaction
in gel immunodiffusion tests with PRSV in sap and WMV-2 antiserum but
PRSV gave a reaction of serological identity when compared with the
Florida isolate of WMV-1.

MATERIALS AND METHODS

Source of Virus Isolates

The isolates were subcultures of those used in a previous study
(Purcifull and Hiebert, 1979). The Florida strain of WMV-1 (WMV-1
Florida) was obtained originally from W. C. Adlerz. The Florida strain
of WMV-2 was isolated from watermelon in Alachua County by D. E.
Purcifull. The Jordanian isolate of WMV (WMV-1 Jordan) was obtained
from G. Martelli, and the Moroccan isolate of WMV (WMV-M) was received
from B. Lockhart.

Maintenance and Propagation of Virus Isolates

Watermelon mosaic viruses, WMV-1, WMV-2, and WMV-M were propagated
in Cuaurbita pevo L. var. Small Sugar Pumpkin. WMV-2 was also prop-
agated in Niooticma benthcaniana Domin. All watermelon mosaic virus
isolates were maintained in separate screened cages. Powdery mildew
was a serious problem on pumpkin, and it was controlled by weekly
sprayings with Dinocap. Frogeye spot, caused by Cevcosvova niaot-ianae
Ell. and Ev. , was controlled on N. beniham-iana by spraying with Benomyl
when symptoms appeared. Pesticides were used at concentrations
recommended by their manufacturers.

Five to six seeds of pumpkin were sown per six inch pot. Niaotiana
bentkaniana were seeded in Jiffy pots consisting of peat moss and were transferred
to six inch pets after 30 days. Seedlings were germinated in a greenhouse from

20

21

which known virus-infected plants were excluded. About one week after
planting and just prior to virus inoculation, pumpkin seedlings were
transferred either to screened cages or greenhouse benches. Pumpkin
seedlings in the cotyledonary stage were mechanically inoculated with
either WMV-1, WMV-2, or WMV-M. Inocula were prepared by grinding
infected pumpkin leaf tissue in water with carborundum. One month old
seedlings of N. benthamiana were mechanically inoculated as above with
WMV-2 infected pumpkin tissue. After inoculation, seedlings were
routinely fertilized every two weeks with a 20-20-20 soluble fertilizer
until harvested.

Virus and Inclusion Purification

Extracts from plants infected with WMV-2 were clarified either with
n-butanol or with a combination of chloroform (CHC1-) and carbon tetra-
chloride (CC1,) (Lima, 1978; Lima et at., 1979). WMV-1 and WMV-M were
unstable in n-butanol and, therefore, extracts containing them were
clarified only with the CHC1--CC1, combination (Lima, 1978). System-
ically infected pumpkin leaves were harvested 21 to 45 days after
inoculation, whereas infected N. benthanriana leaves were harvested 30 to
90 days after the plants were inoculated. The leaves were kept in
plastic bags up to four days at 4 C before purification of the virus.

n-Butanol Clarification Method

One hundred to 700 g of leaf tissue were homogenized in a blender
with two parts (w/v) of a buffer (homogenization buffer) consisting of
0.5 M potassium phosphate, 0.01 M Na„EDTA, and 0.5% Na SO (pH of the
miy :ure was 7.5 to 7.7). The homogenate was filtered through two leaves

22

of cheesecloth and centrifuged at 13,200 g (max) for 10 min. The
supernatant, containing virus, was decanted from the pellet, which
contained inclusions. The inclusions were purified from the pellets
as described below. The supernatant was stirred while n-butanol (8 ml
per 100 ml supernatant) was added slowly. The mixture was stirred for
4 hr at 4 C. The coagulated material was removed by a low speed centri-
fugation at 13,200 g for 10 min. The virus was precipitated (Hebert,
1963) from the aqueous phase by adding 8 g of polyethylene glycol MW
6000 (PEG) per 100 ml of supernatant at 4 C with stirring until the PEG
dissolved, followed by centrifugation at 10,400 g for 10 min. The
pellets were resuspended in 0.02 M potassium phosphate, 0.01 M Na^EDTA,
pH 8.2 (virus buffer) and subjected to equilibrium density gradient
centrifugation in CsCl (p = 1.28 g/ml of virus buffer) in a Beckman
SW 50.1 rotor at 150,000 a (max) for 13 to 18 hr. The virus zone,
located 8 to 11 mm from the bottom of the tube, was collected dropwise
through a hole punched in the bottom of the tube. The collected zone
was then diluted with three volumes of virus buffer. The virus
preparation was further clarified by a low speed centrifugation at
12,000 g and then precipitated as before with PEG. The final pellet
was resuspended in a buffer consisting of 0.02 M Tris, 0.01 M Na„EDTA,
pH 8.0 to 8.2. The virus concentration was determined spectrophoto-
metrically using an extinction coefficient of 2.4 per cm for a 0.1%
solution at 261 nm (Purcifull, 1966). The optical density readings
at 260 and 280 nm were corrected for light scattering by least squares
linear regression analysis of readings taken at 320, 330, 340, 350,
and 360 nm.

23

Cylindrical Inclusion Purification

The first 13200 g pellets from the n-butanol purification method were
resuspended in 100 to 300 ml of the homogenization buffer and emulsified
in a blender with an organic solvent consisting of a 1:1 solution (v/'v)
of chloroform (CHCl-) and carbon tetrachloride (CC1,). The ratio of
organic solvent to homogenization buffer was 1:3 (v/v) . The emulsion
was broken by centrifugation at 4,080 g for 5 min. The aqueous phase
consisting of inclusions was decanted and filtered through glass wool
and the inclusions were precipitated by centrifugation at 16,300 g for
10 min. The inclusion pellets were resuspended in 0.05 M potassium
phosphate buffer, pH 8.2, containing 0.5% 2-mercaptoethanol (2-ME) .
Triton X-100 (TX-100) solution was added to the inclusion preparation
to yield a final concentration of 5% TX-100. This was stirred for one
hr at 4 C. The inclusions were pelleted by centrifugation at 17,300 g
(max) for 15 min and resuspended in 0.02 M potassium phosphate buffer
containing 0.1% 2-ME, pH 8.2 (inclusion buffer). The resuspended
inclusions were layered on a fresh sucrose step gradient consisting
of 6 ml of 80% sucrose, 10 ml of 60% sucrose, and 10 to 12 ml of 50%
sucrose. All sucrose stock solutions were made up in 0.02 M potassium
phosphate buffer, pH 8.2. Inclusions were centrifuged on the fresh
sucrose gradients at 21,000 rpm for 1 hr in a Beckman SW 25.1 rotor.
Inclusions were found layered on top of the 80% sucrose and were
collected dropwise. The inclusion fraction was diluted with 3 volumes
of inclusion buffer and precipitated by centrifugation at 17,300 g
(max) for 15 min. The pellet was resuspended in 0.02 M Tris-HCl, pH
8.2, and the inclusion yield was estimated spectrophotometrically
after being dissolved in 1% sodium dodecyl sulfate (SDS) (Hiebert et al.,

24

1971). Inclusions were examined in a Philips 200 electron microscope
after negative staining with 2% aqueous uranyl acetate, ammonium
molybdate, or 1% phosphotungstate. The inclusion preparations were
either used immediately for immunization of rabbits or stored by one
of the following methods: freezing with 0.02 M Tris-HCl buffer, pH
8.2 at -20 C, freeze drying in Tris buffer followed by storage at
-20 C, solubilization by dissociation in either 1% SDS or in the
Weber-Osborn dissociation buffer (Hiebert and McDonald, 1973), followed
by freezing at -20 C. Inclusion preparations were tested for proteolytic
degradation and contamination with viral capsid and host proteins by
analytical polyacrylamide gel electrophoresis (PAGE) and by serology
against the corresponding virus antisera.

Chloroform-Carbon Tetrachloride Clarification Method

One hundred to 700 g of leaf tissue were homogenized in a blender
for 1 min with 2 parts (w/v) of the homogenization buffer. The
homogenate was then emulsified by adding to 3 parts of the homogenate,
1 Dart (v/v) of a 1:1 solution of CHCl, and CC1, and blending at
high speed for 2 min. Several hundred milliliters of shaved ice were added
during emulsif ication. The emulsion was broken by centrifugation at
480 j for 5 min. The aqueous phase was decanted, filtered through
glass wool and centrifuged at 12,100 g for 15 min. The supernatant
containing virus was decanted from the pellet containing inclusions.

The supernatant was made 1% (v/v) in TX-100. The mixture was
stirred for 1 hr and centrifuged at 12,100 g for 10 min. The virus
was precipitated from the supernatant with PEG as described previously.
The remainder of the virus purification was essentially the same as

25

that described for the n-butanol purification scheme except that an
additional PEG precipitation was often required prior to density gradient
centrifugation. When Cs-SO, was substituted for CsCl during density
gradient centrifugation of WMV-M the amount used was 10.5 g dissolved
in 27 ml of virus buffer and adjusted to a final pH of 8.2 with 1 M KOH.

Purification of the inclusion pellets was identical to the pro-
cedure described previously for the n-butanol purification scheme
beginning with the solubilization step employing TX-100.

Purification of Cylindrical Inclusions for Peptide Mapping

The procedure required 150-200 g of pumpkin leaves showing strong
mosaic and distortion symptoms and free of powdery mildew. The tissue
was homogenized at 4 C in 3 parts of homogenization buffer and 1 part
(v/v) of a 1:1 chloroform-carbon tetrachloride mixture (see virus purifi-
cation). The emulsion was broken by centrifuging at 480 g for 5 min.
The supernatant was filtered through glass wool and the pellet resuspended
in about 250 ml of homogenization buffer and homogenized a second time
at 4 C. The emulsion was broken as before and the filtered supernatant
was combined with the first supernatant and centrifuged at 12,100 g for
20 min. The pellets were resuspended with the aid of a glass tissue
homogenizer, in 36 ml of a pH 8.2, 0.02 M potassium phosphate buffer
containing 0.1% 2-mercaptoethanol (final concentration^ Four milliliters of
20% TX-100 (v/v) were then added and the mixture was stirred at 4 C for
90 min. After centrifugation at 27,000 g (max) for 15 min, the green
supernatant was discarded and the light green pellet was resuspended
and washed in buffer two or three times to remove all green pigments.
Occasionally a second exposure to 2% TX-100 was required. After several

26

washes, the gray inclusion pellet was resuspended in 4 ml of buffer and
homogenized in the micro-homogenizer of a Sorvall Omnimixer at high
speed for 2 min at 4 C. The inclusions were centrifuged at 270 g for
5 min. The supernatant was decanted and a soft pellet consisting
primarily of inclusions was removed by gently resuspending it in buffer
and aspirating. A hard white pellet was discarded. Inclusion protein
concentrations of the supernatant and resuspended soft pellet were
determined spectrophotometrically (Hiebert et at., 1971). The partially
purified preparations were further purified by preparative gel electro-
phoresis using 8% Weber-Osborn gels run in an Ortec 4217 cell with
preparative well former (Ortec Inc., Oak Ridge, Tn.). After electro-
phoresis for 3 to 4 hr, the gels were removed, and immediately
wrapped in clear plastic wrap and chilled overnight at 4 C. The opaque
inclusion band was visible about 2/3 the distance (60 mm) from the bottom
of the gel. The band could be further identified by staining with
Coomassie brilliant blue. The opaque band was cut out by pressing
down on a taut wire held in a coping saw. The band was then cut into
slices approximately 1 mm thick and eluted in 5 to 7 ml of water over-
night at 4 C. The eluate was then removed and saved and the bands
again eluted in water for about 3 hr at room temperature.
The eluates were pooled and small fragments of polyacrylamide
gel were removed by passing the eluates through a 0.045 urn pore size
membrane filter. The eluates were then dialyzed exhaustively against
0.02 M Tris-HCl, pH 8.0 or water. Concentrations of inclusion proteins
were determined spectrophotometrically prior to pooling and again before
freeze drying. The inclusion samples were tested serologically for
antigenic specificity by double immunodiffusion tests. The dissociated

27

inclusions were also examined on Weber-Osborn (1969) or Laennnli (1970)
analytical polyacrylamide gels to test for the presence of host or viral
capsid protein contaminants. After freeze drying the inclusion proteins
were stored at -20 C.

Cleavage of WMV Capsid and Cylindrical Inclusion Proteins

Capsid proteins of each virus were cleaved on the carboxyl side of
methionyl peptides with cyanogen bromide (CNBr) (Gross and Witkop, 1962;
Gross, 1967). The purified viruses were resuspended in either 0.02 M
Tris, pH 8.2 or 0.02 M Tris, 0.01 M Na2EDTA, pH 8.2, at concentrations
between 1-7 mg/ml, and then were cleaved by an excess of CNBr (10-20
mg) in 70% formic acid. The reaction was carried out at room temperature
for at least 18 hr. The cleaved peptides were concentrated by lyophili-
zation, resuspended in 2 ml of water, and stored at 4 C. Each uncleaved
virus preparation was examined on analytical Weber-Osborn gels just
prior to CNBr cleavage to insure that the preparation contained only
undegraded virus capsid proteins (Hiebert and McDonald, 1976; Hiebert
et at., 1979).

Partial proteolytic digestion and analysis on polyacrylamide gels
of purified inclusion proteins dissociated in SDS was performed using
Staphylococcus aureus V-8 protease (Miles Laboratories) as described by
Cleveland et at. (1977). The method involved a slight modification of
the procedure described by Cleveland et al. (1977). The acrylamide con-
centration of the stacking gel was increased from 3% to 5.6% for use
on 1.5 mm thick slab gels.

Conditions were optimized based on a kinetic study in which
purified inclusions at concentrations of 300 jj g/ml were treated by

28

incubation at 37 C with either no protease or 25 yg/ml of S. aureus
protease for times ranging up to 150 min. Proteolysis was stopped by
adding SDS and 2-ME to final concentrations of 2% and 10%, respectively.
The samples were heated at 100 C for 2 min. Fifteen, 30, or 60 yg of
partially digested inclusions were added to each well. Electrophoresis
was performed in a slab gel apparatus (Studier, 1973) at 100 volts, 21
mA for 4-6 hr. Gels were stained with Coomassie brilliant blue and photo-
graphed. Based on these kinetic studies, the standard conditions used
were 15 yg of inclusion protein (per well) which had been incubated for
30 min at 37 C with 25 yg/ml of S. aureus protease.

Polyacrylamide Gel Electrophoresis of
Viral and Inclusion Proteins

Polyacrylamide slab gel electrophoresis (PAGE) was performed
according to the method of Weber and Osborn (1969) as modified by
Hiebert and McDonald (1973).

Gel Electrophoresis of Cyanogen Bromide
Cleaved Viral Capsid Proteins

WMV capsid proteins which had been cleaved by treatment with
CNBr were electrophoresed on a low pH gel modified from Mauer's gel
system #7 (1971). System #7 required three stock solutions: stock
buffer #17 had a final phase of 4.3 and consisted of approximately 48
ml of IN KOH and 17.2 ml of glacial acetic acid plus sufficient water
to bring the total volume to 100 ml. Stock buffer #19 had a final
pH of 6.7 and consisted of 48 ml of IN KOH, 2.87 ml of glacial acetic
acid and was brought up to a total volume of 100 ml with water. The
electrode buffer stock had a pH of 4.5 and consisted of 31.2 g of

29

3-alanine, and 8 ml of glacial acetic acid brought up to one liter.
The electrode buffer was a 10% aqueous solution of the electrode
buffer stock.

Thirty milliliters of the running gel body consisted of 11.75 ml
of water, 3.75 ml of buffer #17, 12.5 ml of the acrylamide stock (30 g
monomer and 0.8 g Bis in 100 ml) and 0.2 ml of TEMED. The gel solution
was degassed for 5 min, then 1.8 ml of freshly prepared ammonium
persulfate solution (15 mg/ml) was added and the mixture was poured to
a height of about 72 mm in an Ortec casting stand. The gel was over-
laid with a solution consisting of 0.3 ml of stock #17 and 2.7 ml of
water. Polymerization occurred after transferring the casting stand
to a water bath set at 40 C. The casting stand with the polymerized
gel was removed from the hot water bath and allowed to cool to room
temperature. After cooling, the overlay was removed and a stacking gel
of about 4.5 mm in height was added. The stacking gel consisted of
0.3 ml of buffer #19, 0.4 ml of acrylamide stock, 0.03 ml of TEMED and
1.5 ml of water. To this was added 0.18 ml of ammonium persulfate and
the gel was mixed rapidly and poured on top of the running gel. The
stacking gel was immediately overlaid with a solution consisting of
0. 3 ml of buffer #19 and 2.1 ml of water. The cap and well gel
solution was composed of 1 ml of undiluted electrode buffer stock,
2.7 ml of acrylamide stock, 6.3 ml water and 0.01 ml of TEMED. The
combined gel was divided into two 5 ml portions for the well and
capping gels, respectively. Polymerization of both well and capping
gels was achieved by adding 0. 7 ml of ammonium persulfate just prior
to pouring. A twelve tined comb was used to form the wells.

30

The sample buffer was made fresh each time and consisted of 1.08 g
of ultrapure urea (Schwartz/Mann), 0.25 g of sucrose, 0.25 ml of #19
buffer and 0.025 ml of a 1% solution of methylene blue. The sample
buffer was brought up to a total volume of 2 ml with water. Approxi-
mately 50 yg of each CNBr cleaved and dissolved protein was dried on a
spotting plate and redissolved in about 30 yl of sample buffer. Samples
were immediately transferred to the appropriate wells. The capping gel
was pipetted into the wells and the gel former was transferred to the
Ortec vertical slab electrophoresis apparatus, Model 4010,4011. The
anode was attached to the upper tank and the cathode was attached to
the lower tank. Electrophoresis was performed for 3 to 4 hr at 200
volts, with a pulsed power of 300 pulses per second and a current of
approximately 70 mA. After electrophoresis, the gel was stained over-
night in a solution containing 50% methanol, 10% glacial acetic acid
and 0.1% Coomassie brilliant blue stain R 250. The gel was destained
in several changes of a solution consisting of 10% methanol and 7%
glacial acetic acid. The gels were photographed and then scanned using
an automatic gel scanner attached to a Beckman Model 25 recording
spectrophotometer.

Serology

Anitsera for virus and cytoplasmic inclusions were obtained by
injecting adult New Zealand white rabbits with either undegraded virus
capsid or cylindrical inclusion protein or with SDS dissociated capsid
or inclusion proteins obtained from preparative polyacrylamide gels.
The WMV-M capsid immunogens used in the production of antisera were
checked by analytical Weber-Osborn polyacrylamide gels to insure that

31

partial proteolytic cleavage (Hiebert and McDonald, 1976) had not occurred .
All rabbits were bled for normal serum prior to immunization. The
initial immunizations consisted of intramuscular injections of 1 to 2 mg
of protein in 1 ml 0.02 M Tris buffer, pH 8.2, emulsified with 1 ml
Freund's complete adjuvant (Dif co) . Rabbits were boosted with one or
two subsequent intramuscular injections two to four weeks apart using
similar quantities of protein emulsified in Freund's incomplete adjuvant
(Difco). All other immunizations involved essentially the same procedure
except that at least one toepad was injected with 0.15 to 0.2 mg of
protein. Antisera to formaldehyde fixed antigens were obtained with
only a single toepad and intramuscular injection. Rabbits were bled
according to the procedure of Purcifull and Batchelor (1977) at approxi-
mately weekly intervals and for three to nine months beginning 10 to
15 days after the first injection.

Antisera also were made against formaldehyde fixed viruses. Dur-
ing virus purifications, antigens were fixed in 1.8% formaldehyde for
10 min at room temperature prior to equilibrium density gradient
centrifugation in CsCl, and prepared for immunization as with unfixed
viruses.

Initially, antisera were freeze-dried and stored at room tempera-
ture. After one year a loss in titer was detected in these antisera
versus frozen antisera. Therefore, the freeze-dried sera were sub-
sequently stored at 4 C or -20 C, and all new serum collections were
frozen at -20 C.

Cross-Absorption of Antisera

Some antisera reacted with healthy plant antigens forming non-
specific precipitates in SDS-double immunodiffusion tests. These

32

antisera were cross-absorbed with concentrated healthy plant antigens
according to the method of Purcifull et at. (1973). Sixty grams of
frozen SSP were thawed and homogenized in 120 ml of 0.1 M potassium
phosphate buffer, pH 7.4, which contained 1% Na?S0_. The homogenate
was frozen for 3 hr, thawed, and centrifuged at 27,000 g for 10
min. The supernatant was centrifuged at 250,000 g (max) for 3 hr in
a Beckman Ti 60 rotor. The resulting pellet was resuspended in 4 ml
of 0.02 M Tris buffer, pH 7.4. Concentrated host antigens were
combined with the antisera to be cross-absorbed according to the
method of Purcifull and Zitter (1973) . Host proteins were mixed with
antisera (1:4, v./v) and incubated overnight at 4 C. The mixture was
centrifuged at 81,000 a for 1 hr in a Beckman Type 40 rotor and the
supernatant containing antisera was used immediately or frozen at
-20 C.

Serological Tests

Double immunodiffusion tests in agar gels, microprecipitin tests,
and enzyme linked immunosorbent assays (ELISA) were preformed during
parts of this study.

Double immunodiffusion tests (Ouchterlony , 1962) were carried out
in agar gels consisting of 0.8% Noble agar (Dif co) , 0.5% SDS (Sigma),
and 1.0% sodium azide (NaNO (Sigma) in deionized water (Purcifull and
Batchelor, 1977), or in a medium consisting of 0.8% Noble agar, 0.2%
SDS, 0.1% NaN3, and 0.7% NaCl in deionized water (Tolin and Roane,
1975). Reactant wells (7 mm in diameter) were arranged in a hexagonal
array produced by an adjustable gel cutting device (Grafar Corp.,
Detroit, Mich.) with a spacing of 4.5 to 5 mm between wells. Plant

33

tissues were extracted by triturating with a mortar and pestle 1 g of
tissue in either 2 ml of deionized water or in 1 ml of deionized water
followed by the addition of 1 ml of 3% SDS. The extracts were expressed
through two layers of cheesecloth. For routine tests, sap from 5 to 10 g
of tissue were prepared in this manner and frozen in 3 to 4 ml aliquots.
When titering an antiserum, either normal serum or bovine albumin was
used as antiserum diluent (Purcifull and Batchelor, 1977).

Sometimes serological distinctions were demonstrated by the intra-
gel cross-absorption technique (Lima, 1978; Lima et at., 1979). Heterolo-
gous or homologous antigens were added to the center wells of a hexagonal
array. The peripheral wells were cut but the agar was not removed until
antigens in the center well had diffused into the agar. After 16 to
18 hr any remaining fluid in the center well was removed by aspiration
and the agar in the peripheral wells was removed. Appropriate anti-
serum and antigens were then added in the usual manner to the center and
peripheral wells, respectively.

Microprecipitin tests were sometimes used to titer antiserum
according to the procedures of Ball (1974) , except that uncoated plastic
Petri dishes were used instead of Formvar-coated glass Petri dishes.

Fractionation of Gamma Globulin for ELISA

Enzyme linked immunosorbent assays (ELISA) were carried out using
a modified procedure of Clark and Adams (1977). The gamma globulins
(yG) used to coat microtiter plates and to conjugate with alkaline
phosphatase were" fractionated from antisera collected at least 2 months
after the initial immunization as recommended by Koenig (1978) . One
milliliter of antiserum was diluted with 9 ml of deionized water and

34

stirred at room temperature while 10 ml of a saturated ammonium sulfate
solution was slowly added. The precipitated gamma globulin fraction
was allowed to incubate at room temperature for. 30 min and was then collected
by centrifugation at 10,000 g for 10 min. The pellet was resuspended in
2 ml of half strength phosphate buffered saline (PBS), pH 7.4, having
a IX concentration per liter of 8.0 g NaCl, 0.2 g KH9P0,, 2.51 g Na0HPO,-
7H?0, 0.2 g KC1, and 0.2 g NaN,. A 10X PBS stock solution was maintained
at room temperature. The yG was dialyzed 3 times (approximately 4 hr
each) against 500 ml of half strength PBS, and then filtered through
a 5 ml DEAE Sephacel column previously equilibrated with half strength
PBS. The vG was washed through the column with half strength PBS, and
collected in 1 ml fractions in siliconized glass tubes. Each aliquot of
the first protein to elute was read on a Beckman Model 25 recording
spectrophotometer. Those tubes containing 0.8 O.D. or more were
combined and the yG concentration adjusted to 1 mg/ml (1.4 O.D.„on)
with half strenth PBS (Clark and Adams, 1977). The yG was stored at
-10 C in silicone- treated glass tubes.

Conjugation of Alkaline Phosphatase
with Gamma Globulin

A crystalline suspension (usually 2.5 mg) of alkaline phosphatase
(Sigma No. P4502, 1000 units/mg), in (NH,)9S0, was centrifuged at 10,000
g for 10 min. The pellets were dissolved in 1 ml (1 mg) of purified yG
and dialyzed 3 times against 500 ml of half strength PBS. A 25%
(v/v) solution of glutaraldehyde was added to the mixture to yield a
final concentration (v/v) of about 0.1%. The solution was incubated
at room temperature for 4 hr during which a slight yellow-brown color

35

developed. The conjugated yG was then dialyzed 3 times against half
strength PBS as described previously. Five milligrams of bovine serum
albumin was then added per milliliter of conjugate and the mixture stored at
4 C.

Preparation of ELISA Plates

Two hundred microliters of coating buffer containing purified gamma
globulin was added to all except the peripheral wells of round bottom
microtiter plates (Cook MicroELISA Substrate Plates #1223-24, Dynatech
Labs, Inc.). The coating buffer contained 1.5 g Na2SO,,, 2.93 g NaHCO ,
and 0.2 g NaN„ per liter and had a pH of 9.6. The optimum concentration
of yG had to be determined for each conjugate, but it was usually in
the range of 1 to 2 yl/ml. The plates were usually incubated 2-6 hr at
room temperature though overnight incubation gave similar results.
Plates were washed three times with PBS-Tween (PBST) which consisted
of the PBS buffer to which had been added 0.5 ml of Tween 20 per liter.
The PBST was dispensed from a wash bottle with great care to prevent
contamination from one well to the next, particularly during all first
washes. Wells were filled with PBST and after at least 3 min the PBST
was removed by shaking the wash solution into a sink. Plates were
blotted on paper towels and washed two more times.

The test antigens were made by triturating leaf tissue in a
buffer consisting of PBST plus 2% (w/v) polyvinyl pyrrolidone (Sigma
PVP-40) , using 9 ml buffer per g tissue. Two hundred microliters of each
test antigen usually were added to duplicate wells. Plates were
incubated at 4 C overnight or at 37 C or room temperature for 4-6 hr .

36

The optimum concentration of enzyme-labelled conjugate was determined
for each conjugate.

Plates were washed 3 times with PBST, and 250 ul of freshly-
prepared substrate buffer containing 0.6 to 1.0 mg/ml of p-nitrophenyl
phosphate (Sigma //104-105 tablets) was added to each well. Each liter
of substrate buffer contained 97 ml diethanolamine, 0.2 g NaN„, and
sufficient HC1 to give pH 9.3. The plates were incubated at room
temperature for 30 to 60 min, and the reactions were stopped by the
addition of 50 ul of 3 M NaOH to each well.

Reactions were assessed visually by rating on a 0, + , ++, +++
scale and by measurement of absorbance at 405 ran with a Beckman Model 25
spectrophotometer. Absorbance measurements required a dilution (1:2)
with water. Results were sometimes recorded photographically, using
backlighting and Kodak Verachrome 64 Film.

The determination of optimum concentrations of coating gamma
globulin, test and control samples and dilutions of enzyme- labelled
conjugates was similar to the general procedure except that the concen-
trations and setups were as shown in Figure 1.

U

CD

C

0

>1

CD

rH

0

h

— 1

T)

(X

o

■u

rH

u

a

C

cfl

n

cfl

cd

n

'J

H

C

CO

B

CO

CJ

U

•-<

4-1

en

(U

-CO) fl 3

0) 1) 4-1 U

a

,n

4-1 -H

T3

m

_rj

m

O 3

S 4-1

jct

4-J

4-1

p.

4= O

ca ca

u

0

c

a. ^h

T1

3

0 rH

" cn

X

(3

en

!-i 0

C i-H

CJ

cn

a

*

4-J 4-1

•H o

2

n

■n

CJ

H i-4

cn

LT

H

tJ

CD CD

3 4-1

—

4J

CO

ax;

X c

CD

<d

ctt

TJ

en u

O O

(-4

Q

U

<

r-i CJ

CD

O

4j

T3 4-1

00

3

C

■n

CD 0

1 OJ

4J

u

G

c

>-JZ

cn

ct)

CJ

en

■h cn

u

c

c

e c

to

o

(U

0

^

u c

H rC

•H

CJ

CJ

M

CD -H

•H 4-1

4J

en

cO

4-1 4-1

4-1 -H

CO

CO

CD CO CO 3 U X*

3 CO

3 a.

w T] ^ O

4-1 O U 1) "4JOX

p<

H

CD

CJ

CD

cO

(3

ffl

0

tj

3

3

4-J

M

o

O

0

cn

CJ

0

CJ

Si.

cr,

CJ

su

j-j

S-'

C

3

cn

CD

CN

■H

0

4J

■!-)

CD

>i

N

C

C

■H

cn

en

JZ

— 1

n

•H

•H

u

u

C)

00 r-l

Tl

4-1

H

cn

■*

a

•H

CI

a)

!-4

3

u

0

JZ

CD

JZ

o

01

-C!

4-1

r-l

Tl

T-1

CJ

c 3

4-1

0

en

•OJ

CD

H

en

a

H

CD

CD

rH

^

0

nH

■a

511
1

CJ

a

4-J

0)

4-1

s-l

II

M

>~

0

X

CD

C

a

c

ej

60

cn

>

CJ

ert

i/-

4-1

TJ

S

r-l

m

•H

C

u

•H

so

4-1

cn

<

.J

^H

3

4J

CD

en

n

a

N

— i

e

CJ

t

-C

u

H

o

CD

CU

•H

CD

P^

:j

4-1

a

_C

X

4-1

H

N

•H

c

H

-n

c;

en

Q.

.1)

X

OJ

cfl

cn

CO

H

o

to

1)

3

CJ

cn

X

CD CO •

6 cd cn
cd e '-<

38

I

=> <

iinwm ■»i»io»iii»*iimi

imriiii ■ nwt tyuwir--
I f

1

n

WWW KawWMj WMWIWI mwKSi

I!

! i

mn»MPimw.,iwiiiii< nwt mm is;

OOZ^T

008 -T

3

ooz;:t

—
C
O

006£=T

•a

QJ

008 = 1

X5

r P

00^1

003e=i

4-1

O

008 = 1

c
o

•H
4-1
3
H
•H

a

002 = 1

u

(X

w a

w

S

c e

0J

TO

O cr)

H

CO

U CO

RESULTS

Purification and Properties of Watermelon Mosaic
Viruses and Inclusions

Purification schemes for virus and virus- induced inclusions are

shown diagramatically in Figures 2, 3, and 4. Virus yields were

determined using an extinction coefficient for TEV of E ' ° = 2.4

261

(Purcifull, 1966). The typical ultraviolet absorption curve (Figure 5)
obtained for all purified WMV isolates had a maximum absorbance between
260 and 262 nm and a minimum at about 245 nm. The 260/280 ratio was
approximately 1.2 after correction for light scattering. This value
is consistent with those found for other potyviruses {e.g., Lima, 1973).
WMV- 2 infected pumpkin or N. benthamiana tissue yielded approximately
10 mg of virus per 100 g fresh weight when n-butanol (Figure 2) was
used in the clarification process, as opposed to only 3-4 mg per 100 g
fresh weight when a 1:1 (v/v) mixture of carbon tetrachloride-chlorof orm
(Figure 3) was used. Both the Florida and Jordan isolates were unstable
in 7-i-butanol and pumpkin infected with these viruses routinely yielded
approximately 10 mg virus per 100 g fresh weight with the carbon
tetrachloride-chloroform clarification method. WMV-M, also unstable
in 7Z-butanol, gave maximum yields of only 3-4 mg per 100 g fresh weight
with the carbon tetrachloride-chloroform method.

To prevent irreversible aggregation of WMV-2, it was necessary to
add 0.01 M Na„EDTA (final concentration) to each solution throughout
the procedure. Neither WMV-1 nor WMV-M required Na„EDTA though it was

39

Figure 2. Flow diagram outlining the procedure of purification
of WMV-2 using n-butanol as the clarifying agent,
polyethylene glycol (PEG) for virus concentration,
and CsCl equilibrium density gradient centrifugation
for separation of virus from host components. See
description in Materials and Methods for further
details.

41

PELLET —
(discard)

PELLET —
(discard)

SUPERNATANT-
(discard)

SYSTEMICALLY INFECTED TISSUE

0.5M KP04 pH 7.5 + 0.01M Na0EDTA +0.5% NaoS0~

HOMOGENIZATION

FILTER

I
CENTRIFUGATION: 14,600 a-10min

SUPERNATANT

I
8% n-BUTANOL

\
STIR 6hr

I
CENTRIFUGATION: 10,400 a-lOmin

SUPERNATANT
I

FILTER

I
8% PEG

I
STIR lhr

I
CENTRIFUGATION:

16,300 -7-15min

PELLET

0.02M KPO, + 0.01M Na0EDTA pH 8.2

I 4 2 r

CsCl GRADIENT CENTRIFUGATION
d=1.28 g/cc - 120,000 g max-15-18hr

COLLECT VIRUS ZONE

I
DILUTE 3X WITH 0.02M KPO, + 0.01M NanEDTA pH 8.2

| 4 2^

CENTRIFUGATION: 12,000 a-lOmin

PELLET

(discard)

SUPERNATANT

8% PEG
1

STIR lhr

CENTRIFUGATION
|

SUPERNATANT

H

(discard)

i

PELLET
1

0.12M TRIS

0.01M Na2EDTA V

VIRUS

Figure 3. Flow diagram outlining the procedure of purification
of WMV isolates and the initial separation of the
cytoplasmic cylindrical inclusions from the virus.
The procedure uses chloroform and carbon tetra-
chloride as the clarifying agents and a low speed
centrifugation to pellet the inclusions (see
Figure 4) . Polyethylene glycol (PEG) is used to
concentrate the virus, Triton X-100 (TX-100) for
solubilization of the pigments, CsCl or CS2SO4 for
separation of the virus from host components. For
detailed description, see Materials and Methods.

SYSTEMCALLY INFECTED TISSUE

i

0.5H KPO, pH 7.5 * 0.51 Na„SO, (* Q.01M Na,EDTA for WMY-2)

CHC13 ♦ CC^

lOMOGENIZJUION

I
CENTRIFUGATICN: 550xg-5m1n

=ELL£T ! SUPERNATANT

I
3.5M <P0, * 3.51 NSoSO, -ILTER

;- Q.G1M Na,£DTA for *!«IV-2) |

•omogenization -^-combine supernaiants

I I

CE.NT3IRJGATICN: 5SOxa-5min ; CENTP.IR'GATICN: ] 5,3COxg-; Smin

j SUPERNATANT FILTER > j PEU.E7 ( Inclusions) I

(See Next Figure) .
3ELL£T
(discard) SUPERNATANT (Virus)

35 PEG

STIR Ihr

CENTRIFUGATION: iO,400xq-!Cimn

JPERNATANT i

[discard) PELLET

SUPERNATA.HT-

0.C2.M <PO, (* 0.Q1M Na,£3TS £or JMV-2!
pn 3.2 - "i -RITCN-<1CQ

STIR Ihr

CENTRIrllGATICN: 12,3COxc-;0 inn

SUPERNATANT

3i PES

STIR 30nin

CEHTP.irJGATION: 'C,W0xg-1Cmin

SUPERNATANT

(discard) PELLET

3.02M KPO. (+• 3.01M sa-,£2TA for M-Z) ?H 3.2

CsCi GRADIENT CENTRIFUGATICN .ZiSQi 'or *W-M)
d»1.28g/cc - 12C,0OOxg - 15-i3hr

COLLECT VIRUS ZONE

DILUTE 3X C.02H KPOj (+ O.JIM Sa-sDTA for iWV-3)

CENTRIfUSATION: IQ.iOOxg-lOmin

PELLET 1

'discard) SUPERNATANT

8X PEG

STIR SDnin

i
KHTRIFUGATION: I0,400xg-I0m1n

(discard) PELLET

j 0.02J1

TS

$

pH 8.2

I

| (*o.

01«

na

,£DTA

r3r

*<V

2)|

Y

R

",

Figure 4. Flow diagram outlining second stage of purification
of cytoplasmic cylindrical inclusions (for first
stage, see Figure 3). Remaining pigments were
solubilized with Triton X-100 (TX-100) and inclusions
separated by centrifugation. Large aggregates of
inclusions were broken up by homogenization at
9,000 rpm 3 min (max.) in a Sorvall omnimixer micro-
hcmogenizer.

Inclusions to be purified as dissociated protein sub-
units were separated from a hard starchy-like
pellet by very low speed centrifugation. The
inclusions found in the resulting soft pellet and
supernatant were dissociated in 1% SDS and purified
by preparative PAGE. Inclusion subunits were
eluted from the gel and freeze dried.

Whole inclusions were further purified on a
sucrose step gradient, washed, concentrated
and stored at 4 C in 0.02 M Tris buffer, pH 8.2.

45

PELLET (INCLUSIONS)

I
0.05M KPO. pH 8.2 + 0.1% 2-ME

HOMOGENIZATION

5% TRITON- X 100

STIR 90min

I
CENTRIFUGATION: 27,000 a-lOmin

SUPERNATANT-
(discard)

FOR WHOLE INCLUSIONS.

PELLET

WASH 3X 0.02M KPO. dH 8.2 + 0.1% 2-ME

I 4 *

HOMOGENIZATION SORVALL

OMNIMIXER MICROHOMOGENIZER
9000RPM 3min (MAX)

i

SUCROSE STEP GRADIENT

CENTRIFUGATION: 45,000 g-lhr

i

COLLECT INCLUSION ZONE

I

DILUTE 0.02M KPO, pH 8.2 + 0.1% 2-ME

I 4

CENTRIFUGATION: 27,000 a-15min

SUPERNATANT
(discard)

PELLET

0.02M TRIS pH 8.2
INCLUSIONS

.FOR INCLUSION SUBUNITS
I (PAGE purification)

WASH 3X 0.02M KPO, pH 8.2 +
0.1% 2-ME

CENTRIFUATION: 27,000 -7-10min

SUPERNATANT
(discard)

PELLET

!

0.02M Tris pH 8.2

HOMOGENIZATION SORVALL OMNIMIXER
MICROHOMOGENIZER 9000 RPM
3min (max)

I
CENTRIFUGATION: 250 _7-5min

HARD PELLET
(discard

DECANT SOFT PELLET AND SUPER-
NATANT

I

DISSOCIATE IN SDS

I

PREPARATIVE PAGE

I
ELUTE IN 0.02M TRIS

I
DIALYZE 0.02M TRIS

I

FREEZE-DRY

INCLUSION SUBUNITS

Figure 5. Absorption spectra of purified preparations of
WMV-M in 0.02 M Tris-HCl buffer, pH 8.2, and
WMV-M dissociated cylindrical inclusions in
0.02 M Tris, pH 8.2, containing 1% SDS.

47

0.8 •

0.6 L

0.2 r

PURIFIED WM V- M
V! RU S
NCLUSIONS

240

260

280

300

320

340

360

48

usually included. All virus isolates at about 1 mg/ml in Tris
or phosphate buffer showed strong stream birefringence.
Most proteains were not detected by PAGE (Figure 6).

Cylindrical inclusions were unstable in n-butanol and were
always purified using chloroform-carbon tetrachloride for clarifi-
cation of homogenized plant saps (Figure 3) . Yields of up to 40
A280 units (1 A280 = -^ m§ Protein) per 100 g fresh tissue weight
were obtained. A typical ultraviolet absorption is given in Figure
5. Highly purified inclusion proteins were obtained by elution
from preparative polyacrylamide gels. Yields from such gels
averaged about 1 mg purified eluted inclusion subunits for every 3-4
mg of crude inclusion preparation. These highly purified inclusion
proteins which reacted with antisera to untreated inclusions were
used as immunogens and for partial digestion by 3. aureus
protease.

Infectivitv of Purified Viruses

WMV-1 (Jordan) and WMV-2 (Florida) purified by these methods and
freeze-dried with 0.02 M Tris and 0.01 M Na^EDTA at a final concen-
tration of 1 mg/ml were infective to a dilution of about 10 mg/ml
after lyophilization and storage for up to one month at room tempera-
ture. Infectivity of WMV-2 was determined by inoculation to a local
lesion host, C. amarcnticolov , whereas WMV-i (Jordan) infectivity was
determined by mechanical inoculation of pumpkin at various dilutions
and observing systemic symptom development.

C JZ
•H CTn

qj w

a.
g

C 3

o -H

•H TO

M O

3 cn

H _
O V.

a

O

o

o
o

CD

o

o

4-1

C

M

r.

3

CD

CTv

r-»

Vj

0

rH

CN

^3

CD
3

CO

H

0)

c

C

cl

s

Cfl

•H

H

CD

i-H

1

co

s

H

3

M

a

TO

CO

c)

->

TO

rQ

CD

"J

>! M

"3

cn

r->

-C

cr)

CO

a

0

(3

'H

•H

S

u

03

id
3

60

CD
J-J

dj

CJ

M

TO

n

a

•H

CD

)-i

cn

CO

c

CO

4-i

Q.

C

cn

T— 1

O

0

0

•r-l

-n

01

TO

4-J

cn

H

a

QJ

•H

— 1

>,

cfl

cn

•H

a

cn

CO

H

-a

CJ

>

c

fl^

X)

CO

■H

0

CD

CO

3

4J g r- 1

H

P

C

r*4

4J H

c-

0)

„

0

CO

t/3

cn

M

0

0

TO

co*

■H

'J

0^

3

Jl

4-1

r—

H CD

OJ

CO

\0

cn

•H

0

1

ft C

rC

u

X

H

jo

CO

3

|

CO

-J

M

u

Oj

TO

cn

'J

•rH

M rH

a)

3

H

c

' '

00

CD

s

OJ

4J

TO

0

so

•H

00

TO

c

•i-J

r* »

CJ

c

CD

c

C

0

rj cn

CD

2

ct)

M

>,

■r-l

CO

CO

'->

pq

CO C

4J

CJ

T

>,

Cii

S O

■1)

CD

CD

M

•H

00

CJ

r--

4J

TO

~"

U

CO

TO

CO

0

T3

•« rH

CO

QJ

c

—i

•H

>,

•H

CO

CJ CO

0

3

CO

c

X

TO

—1

IH

cn

•H

a

CO

CJ

<

TO

0

CO

i — i

TO

■CJ

a

ai

CO

C O

u

•H

•H

■H

to

o

cn

cfl ©

0)

GO

■CD

en

Jl

r^

CO

V

o

TO

4-1

u

OJ

•H

>

CN

3

H

C

M

CO

0Q

H

CO

1

CO

to cn

C

"Si

u

CO

CJ

CD

>

rH

m

CD

O. rH

|

n

C

CD

or

>

TO

'J

r-

W oj

■H

3

r*

■Si

•H

4J

CO

cn

Z

3

rJ

hM

TO

cn co

TO

U

•H

u

co

CD

0

0

•H

CD c

li

CO

00

Cfl

TO

H

H

U-l

H

c <D

TO

H

3

CO

a)

J»>

■CD

co

o

CO so

CO

3

H

CD

rl

rJ

UH

TO

o

1— 1

iH o

0

CJ

CJ

cm cn

0^

0)

1 C

j3

>

e °

OJ

0

iH H 4-1 CO

Ti

CO

14_

CO

^— ,

y

CO

X

—

■r-'.

CO

CJ

TO

CO

CO

c

B

0

4-1

CJ

Si

c

cn

G

CO

3

•H

rd

cn

:"

-H

>y

•H

—J

4J

0

0

4-1

M

E

CO

3

4-J

r-J

—4

3

r-

4J

CJ

CO

-Si

CO

5

—1

rH

CO

rH

a.

0

r"

co

•H

•H

cO

c

CJ

SO

IH

TJ

V

co

•H

■a)

UH

: — |

-o

mH

■rt

CD

rj

u

CD

4J

0

r4

. ry

rl

4J

CO

D

CJ

4J

c

soo

cn

TO

X

»-3

u

T

3

— 1

c

rJ

c

CO

CO

3

SO

cn

4-

r4

Cj

OJ

•H

a

•H

CJ

CD

•H

3

cn

CD

Cj

!-

4-1

OJ

rH

c

C

cn

'J

0

V

'-

T)

•H

CO

o

u

>,

CO

CO

<;

3

<J

3

0

r4

•H

CJ

cn

cO

a

H

rH

3

r-4

rH

JO

'CD

•-0

~-

•H

CO

H

cfl

CO

^

'■■■

TO

cn

c

^-t

:o

TO

>

Q

t-

r-i

CO

M

C

>c

rO

CO

cn

cn

rH

u

cfl

CJ

■H

0

•H

CJ

to

C

TO

3

O

^J

4-1

e

li

S-4

CD

CJ

0

0

•H

CJ

o

J3

"J

JO

^J

TO

CJ

u

0

-J

4_.

cn

fjj

m

so

a

TO

'J

CO

CO

C

M

H

rH

a rH

■H

i — i

CO

C-

CO

4J

co

CO

U

0

CO

CO

cfl

0

■c

CD

to

CO

^^

to

CO

J2

~^

to

s

ts

TO

u

E

m

3

50

-<&

I

•

C5

1

CM

<*»J

m

51

Particle Length Determination of WMV-M

Measurement of 350 purified WMV-M virus particles gave a normal
length of 713 nm (Figure 7). Approximately 50% of the measured particles
were between 675 and 750 nm. Approximately 5% of the examined rods were
clustered between 1400 and 1440 nm as an apparent result of dimer
formation due to end-to-end aggregation.

Molecular Weight Determination

Molecular weights of capsid and inclusion proteins dissociated in
SDS were determined from 8% polyacrylamide analytical gels (Weber-Osborn,
1969) employing proteins of known molecular weights as standards
(Figures 8 and 9). The undegraded capsid proteins of WMV-1 (Florida
and Jordan isolates), WMV-2 (Florida), and WMV-M, each tested at
least four times, gave an average molecular weight for undegraded
capsid protein of 34,000 daltons with a range from 32,500 to 36,000
daltons. Molecular weights of the cylindrical inclusion subunits of
all four WMV isolates ranged from 68,000 to 71,000 daltons based
on 18 determinations, with an average of 69,000 daltons. WMV-1 and
WMV-2 virus preparations which were stored at 4 C showed typical
proteolytic cleavage with time (Hiebert et at., 1979; Hiebert and
McDonald, 1976) to lower molecular weight forms of approximately 26,000
to 30,000 daltons (Figure 8). Purified preparations of WMV-M did
not have lower molecular weight forms resulting from proteolytic cleavage
on any of the preparations run on polyacrylamide gels. Other than the
apparent resistance of WMV-M to proteolytic cleavage, significant
differences in the molecular weights of the four isolates were not
observed.

Figure 7. Histogram of lengths of WMV-M particles from
purified preparation negatively stained in
phosphotungstate. Normal length was 713 nm
based on measurement of 350 particles.
Class interval is 25 nm.

80

60

LU

_i
O

t-

<

a.

40'

53

cr

Ui
00

20

500

! !
! i

-I 1»

1 !

• I

6 00 - ,700 800

PARTICLE LENGTH (nrn)

200

W M V - M0R0CC!

OJ 01 0) o

J-J 4-1 C t— I

BJ O to i-H

S CO

c

0

4-1

. *

r-t

CO

tO

rj

T3

"i

4J

O

^h

O

CO

m

4J

q

3 -H

o

o o

3 C

c

4-1 o

r— 1 LO

m

y* o

CO -

■H

c >>

4J

x) r~

'J)

O Xj

c

CN

>,

•H

CO

o

-4

4-1 -O

c

o e

•

cfl

CO 01

H

m !-l

01

c

!-i -H

fci

•> O

C

CO

CO l-i

CO

r-^ u-t

O

a, co

4-1

CN

4-1

3

01 >

e x ca

S co

u-i as u (n

•H 01 01

u -in j:

3 C M-l 4-1

a, oi 3 c

O O CO W

CO 01 O 3

>> 4-1
Sj 01

a xi

CO

>. CO

H Jd

O 01

P. 01

3

00 01

£ 4-1

3 O

■a a c j

i3 AS

a u

CO CO

e

O CO O O

ill Hi h in Bi

X) S 01 ro X!

M > „

O 0) -H XI _

O 4J O 00 01 o

C 0)
3 T)

H S tCC 0 3 00

•H U 0) ~4 4J CO 01

i-icflco 13 ^ H T3 "O

XI ,J3 3 4-1 O

••> 4J ftHO

O /-> -H co «

C CO 3 2 XI CN

O C I m

« o en > o

r* 4-i c 2 O S

n h o S o h

■^ CO 4-1 « o

XI H J-i <t U-i

01 CO OJ ro

81 O TJ J „ 01

CO O 4-1 XI 4J

C O O O 01 CO

01 O
UX!C 4-1 00 CT\ m >-, CO

aico « h o cm «cu
ft. u <t co s-i ~— ' <r ca QQ
co oo m xi x! en 0)

01 01 „
U X) XI

>> 01
-3 01

01 CO

O x) s-4 xi 3 i-l

X) CO O -

fO i-H JX, S-l rH CO -

<r O ,3 00 rH XI 3

^4. C 01 OH JJ

o co xi m-i )-i xi

CO X! "H 3

O r-t CN ^i JS 4J

■H

rH

0

i

1

c^

1

1

a

i— 1

*J

U

a)

fc=

>

>

>

cil

CO

01

00

2

^

vO

y1

H

>

S-l

<r

_TJ

H

fS

CN

3

^

pq

O

0

4J

01
XI

4J

•H

1

1

1

1

1

1

ft.

cfl

•H

3

0

id

<

S3

c ;

a

W

ta

u

4-1

X!
01

CO

H

hi

CO

~(

01 fa

h-)

CO

'0

0

—

c

>

3

4-1

■H

-3

jj

0

^J

U <~t

rH

n

CN

rj

4-1

(0

40

1

01 1

1

4J

1

M

cfl

:n

CO

^J

1

CO >

>

H

>

.— i

i-H

01

CO

rCl 2

CO

01

3

.1!

>H

u

3

O 3

X)

2

3

CJ
1-4

■00

C

01 O U 00
rH 4_i O ft
Ed CO cfl U

X> fa! i-l

rH 01 rH

2 U M CJ

u ,e o) xi

01 00 Ur

H H U4 O

O 0) T-t o

2 3 x) m

55

J

I

1

1

21

J

~

U-i

•-N

-i

c

03

en

cn

1)

u

1-1

•H

r-

en

C

3

X

to

01

M

xi

en

■n

R

UJ

r-

c

03

c

M

to

03

CU

a

OJ

u

01

a

01

OJ

en

UJ

01

•H

JO

u

CO

0)

JJ

u

>,

>>

0

e

H

u

-n

H

4J

0

CO

o

c

,o

•H

C

4-1

CO

CO

CU

4-1

«3

.a

3

03

03

<

0

0

CU

c

a

M

-o

60

0

a

OJ

O,

■H

H

4-1

c

>* T3

03

p-l

CO

01

i— 1

en

a

t-*

CO

H

1)

OJ

TD

ss

,.i

CO

CO

M

c

fl

ooo

H

su

0

u

<u

o

■0/

c

™!

4J

^

in

24J

■H

u

o

~

Tl

C

u

M

H

'"J

en

1>

a

cr)

6

'

4J

a)

Cfl

en

H

c

>N

e

CO

en

(J

•H

,£3

—
fa

57

J

3
I

-

1

1 •

J

J
J

58

Capsid Protein Digests by Cyanogen Bromide

Digests of capsid proteins of the four WMV isolates by CNBr (Gross and
Witkop, 1962) revealed at least three distinctive patterns when analyzed by
discontinuous PAGE. WMV-M and the two WMV-1 isolates (Florida and Jordan)
formed similar patterns with Mauer's gel system # 7 (1971) (Figure 10) while
the position of bands from WMV-2 digests gave a consistently different
pattern. All four isolates had two major bands and about seven to eight
minor bands. The digest patterns were similar from gel to gel and with
cleavages from different purifications of the same isolate. The two
major bands, 6 and 7, of WMV-1 Florida showed a slightly slower electro-
phoretic mobility than the two corresponding major bands of WMV-M and
WMV-1 Jordan. Band 9, which was present in WMV-M and WMV-1 Florida, was
missing in WMV-1 Jordan. Band 8, which was detected as a shoulder in WMV-M
and WMV-1 Florida, was not resolved clearly in WMV-1 Jordan. Analysis of
peptide patterns on gels made with buffers having a lower pH than the system
described here suggest that WMV-M is distinct from WMV-1 Florida and
WMV-1 Jordan (data not shown).

Cylindrical Inclusion Digests

Purified cylindrical inclusion proteins eluted from preparative
gels and cleaved by Staphylococcus aureus V-8 protease (Cleveland et at. ,
1977) gave different peptide fragment patterns for WMV-1 Jordan, WMV-2,
and WMV-M when analyzed on Laemmli discontinuous gels (Laemmli, 1970)
(Figure 11). While the digest patterns for the inclusions of each isolate
were distinctive, there was no apparent variation in the inclusion
patterns of the same isolate purified on different dats.

Serology

Antisera specific for both virus and cylindrical inclusions of
WMV-1, WMV-2, and WMV-M were obtained (Tables 3 and 4). Unless

Figure 10. Electrophoretic analysis of cyanogen bromide cleaved
WMV capsid protein. Undegraded capsid proteins were
cleaved by cyanogen bromide and electrophoresed on a
12.5% polyacrylamide gel using a discontinuous buffer
system (pH 4.5) of acetic acid and 6-alanine. Gels
stained with Coomassie brilliant blue were scanned
at 565 nm. Lanes from left to right contain the
following digests:

WMV-1 Florida Lane "1A" and the next lane to

its right
WMV-1 Jordan Lane "J"
WMV- 2 Lane "2" and the next three

lanes to its right
WMV-M Lane "M" and the next lane

to its right.

60

CNBr Cleavage

C.-8-7?

ivIDt RtSUUS
{.LEAVA&E 0* CAPS1D MOTtlNS OF VF*V-N, WW-1 (FlONIDA AMD
t, AND MMV-2 (FL0«IDA> r.AVE CONSISTENT RESULTS F*
HUM DlfffREHi PUBLICATIONS AND CLEAVAGES Of
HO* T IONS ST09ED FON PERIODS Of V
■■ Flopii>a and Jordan ga
I'WDANIAN FNA
LN FRAGHCMTS HAVE

ucs. The piftid

[ IN f khEH COUP.

; J".-".

WMV-2 Florida

WMV-M Morocco (M)

WMV-1 Jordan (J)

WMV-1A Florida lift)

0 2 4 6 8 cm

O CO U 60

c

•H

fH

a

u

m

3

0)

3

J3

-J

— i

M

01

•lj

CJ

^

e

m

3

C

«

p,
u

H

to

1)

0)

g

01

en

<

w

oo

—

3)

rl

0

33

>,

3

•a

■-!

0)

H

0

03 C 01 —I
0) 3 CO > t— I

03 H J i-l O

*J

•H

01

'/)

0

a

to

M

T3 ^3

01

Pi

01 <C

-J

hj

4J Oj

~<

i-i

oo

3) 1

en

1

oj m

cu

>

CO G

u

*j

•H 00

a

r

c >,

3J

•H

.',-.

3 _C

M

-^

_c

~

OJ T3

M

o

':

,fi 01

3

U

<tj -rt C rH

5 3 as
'j e

O 03 01 03 O ••

r-J C S-i 0) U C

3> O 0) C u-1

rC -H 2 CO

fij 03 H 91

a a 03 oi

■UH u « C

0 j U C CO 53

a oi u *j 03

O CJ 4-1 CO C

C3 C -H 03 3
P. -H +J 01 rH

6 h a ec u
o ^ 0) -h c

O CJ P. T3 -H

62

63

6 fe

en

j_i

a

&ti

tJ

0)

u

TJ

tx,

£

»

64

> u

§ 2

<u -a

T3 CD

c« -H

a -u

OJ c

CO CD

O 6

C *->

3 CO

S <D

S-i T3
(X, C

3->

JS

OJ

O

a

4j m

•H

c

•H

a n

33

3

60

o w

£

CD

s

T3

a

£*

u

K

-*

13

fl

X)

3

3

i_l

"3

,•

V

0)

0

0

•H

cu

4J

>.

•H

so

•H

•H

3

^

cn

rH

■H

i-1

iJ

CJ

c

s:

4_i

CJ

CJ

T3

CU

4J

0

i

3

-!

CO

a)

,C

^>

a

CO

CD

:.;

i-i

TU CJ

>1

1)

1-1

i-i

T3

a

<v

t — i

>

§

t^ j=

0)

CO

4-1 4-1

cO

rC

■i-i

CD

OJ

•u

CD

cn o

o

M

S-J

•H

>

>

CO

S-J

cu C

c

"3

3

•H

•H

OJ

4J

s

2

0)

4J

4-1

4-1

s

3

3

3

M

c

£

■H

CO

3

& 01

u

u

11

■H

C

w

to

4J

U

ro <iO

CD

V

4-1

—

•r-i

0

OJ

o

11

cn o

■J)

a

s

u

LJ

ft

3

c

cn

3

•H

JJ

1— 1

D

CJ

•H

^ 2

4-1

■H

c

-l

cO

II

II

II

4-J

1— 1 g

rj

C

w

0)

0)

£

3 6

<

<

•H

Q

«

J-

1

o

<

O H

CO ,a

a "a

JJ u-i M

65

Table 4. Serological reactions with WMV inclusion antisera.

Antisera

WMV-1
Jordan

WMV- 2
Florida

WMV-M

Rabbit No.

Inmiunogen
Treatment

903

None

904

942

943

928

Sub units
None None (PAGE)

9351

Subunits
None (PAGE)

Inclusion
Antigens

WMV-M

WMV-1
Jordan

WMV-1
Florida

WMV- 2

DMV

BCMV

SoyMV

LMV

PRSV

Reaction with Antigens

+

+

+

0

0

+

0

0

+

0

0

+

0

0

+

0

0

0

+

0

+e

0

+

0

-

0

-

0

+

0

+

0

Notes :

Antisera from all bleedings gave strong healthy reactions
Weak homologous reactions to all bleedings

, Subunits purified by SDS polyacrylamide gel electrophoresis (PAGE)
Reaction with antigens in SDS double immunodiffusion tests:
+ = positive reaction
- = negative reaction
0 = not tested
Reacted with purified inclusions, but not with sap.

66

specifically noted, antisera did not give a positive reaction with
healthy antigens in gel immunodiffusion tests. All the antisera gave
positive homologous reactions. In reciprocal SDS double immunodiffusion
tests, reactions were negative between WMV-1, WMV-2, and WMV-M with
virus antisera obtained during the first four months following
immunization (Figure 12). The formaldehyde- fixed WMV-M antisera
reacted with WMV-1 (Florida) and WMV-1 (Jordan) after four months
(Figure 13) . This antiserum was made from a virus which had a capsid
protein molecular weight of approximately 34,000 daltons, indicating
that it had not undergone proteolytic cleavage (Hiebert and McDonald,
1976). In SDS double immunodiffusion tests, the fixed antigen reacted
with WMV-M antisera but not with WMV-1 antisera. WMV-M antisera
(bleedings taken for three months following immunization) made against
nondegraded capsid which was not formaldehyde fixed did not react
with either isolate of WMV-1 (Figure 13). Antisera collected after
four months formed only faint precipitin bands with WMV-M, precluding
further studies of cross-reactivity with WMV-1 isolates. None of the
other antisera to WMV-1, WMV-2, or WMV-M, representing bleedings taken
up to one year after immunization, gave heterologous reactions.

Florida and Jordan isolates of WMV-1 gave reactions of identity
in reciprocal double immunodiffusion tests. Intragel absorption tests
failed to detect any serological differences between these two isolates
(Figure 14) .

In reciprocal SDS double immunodiffusion tests, WMV-2 reacted
with bean common mosaic virus, blackeye cowpea mosaic virus, soybean
mosaic virus (Figure 15). These viruses did not react with WMV-M or
WMV-1 in reciprocal tests.

Figure 12. Reciprocal SDS-double immunodiffusion tests between
WMV-1, WMV-2, and WMV-M, with antisera obtained
during the first four months after initial injection
of immunogen. Media contains 0.8% Noble agar, 0.5%
SDS and 1% sodium azide.

Center wells were charged with:

lvs = WMV-1 Florida antiserum
2vs = WMV-2 Florida antiserum
Mvs = WMV-M antiserum.

Antigens in peripheral wells were as follows:

WMV-1 Florida in sap
WMV-1 Jordan in sap
WMV-2 in sap
WMV-M in sap

corresponding purified virus preparation,
50 yg/ml
H = healthy pumpkin sap.

e

05 tO rH 3

3 C 3 U
O -H -H 01

to 3 u en
O O "H -H

01 0)

X! £

4J

to

3 JS

•H AJ

3 -h
0 3

O T3

w pq C U

6 &

3 a

0) >

- 3

s

3

4J

IH

S-i

I

1

M

1

■n

eg

_3

>

05

E

1)

•u

-J

2

H

■H

■3

I+-I

3

•H

3

i — i

X

>,

3

3

01

C

3

3

U

01

05

u

jj

3

a

T"

c

— '

•r-i

M

'~

3. r-i

-3

g

1)

CO

cd

c

. «

x>

u

4J

£

~

S

3

g

~

*-

£

i

M

14-1

3

V

c

>

0

a

u

4-1

X

s

:-i-i

-J

■H

3

C

>-.

3

rC

a

to

H

(J

•3

3

0

to

H

— 1

>,

3

0

—

tu

0

3

i — i

X

—i

0

4J

u

0)

•H

i-i

n

cd

-j

4J

4-1

3

1/

•H

3

1

N

X

0

05

X2

£

3

■H

05

>

-2

•r-t

—

g

•H

^

cd

X

^

3

0

*— '

ij

c

«

;~

H

M

■I)

X

1

s

u

0-

rTZ

• r-

3

>

3

-d

O, t-H

m

>-i

3

3

H

y-

§

u

E

3

•- 1

a

i+j

TJ

M

•- i

■a

g

■H

c

11

C

i_l

0

U

u

U-,

•H

a

•r-i

C

3

U

3

3

Cd

0

3

■H

i

•r-1

r-i

•ui

£

en

y|

J-J

r-l

•H

U-l

•-H

3

c

1

3

0

3

C

<u

■H

•h

^j

'J

H

ij

3

3

?—s

— i

_C

—

U

05

-3

3

CO

<— ■

g

>

-a

3

3 -H

P.

P.

05 S-i

3

3

3

05

-n

3 3-

t-I

3

3

A

~i

■n

3 3

-^

axj tj

li

&

3 -H -H

d

3

=

05 - -

05

—

O 0

M

a.

3 -i rH

3

3

■H h h

~i

>-l

^

gHH
1 1 1

CM

1

1

S3S

3

II

>

3

3

2 H >

>S

fg

70

JO 9\
© ©

© e

© ©

en s-5

a rH

CO

-a

60 3

3 co

to «

T3

->

3 CO

C

T3

C

Q

CO

C

cfl

CO 00

T3

^

CO

•h r>«

rH

H

u m

B

rH

R

o •

•**.

r-t O

DO

ftf

Pn

e

to

~

*

g

rH rH

h

H

I nj

H

> cO

■•-^

g a

co

R

(Tt

»^»

T3

c

3

"3

a s

6 N

3 w 3 « TJ 3 tI

■a o t) u o T3 s -a cm

H IH CO 4-1 CO rH

-a -h -a i

3 -O a T3

M -H V-l -H

3 *h 3 h

& -a

S -I-1

£X h£ hi j-i

X! "rt x! o

•H

H

•H

—1

■H

•H

r-H

•H

—i

•H

3

1

3

1

3

13

1

3

i

f?

rrj

>

-a

&

Tl

-1

5

ti

2

t:

11

,£

3J

2

01

aj

<U

o

SI

si

Of

M

M

ex

Sh

_3

'h

-3

Sh

M

c

Sh.

JZ\

M

co

H

CO

W

cci

CO

u

CO

u

CO

X!

•H

_c

•H

JZ

rC

•H

JZ

•rH

CJ

3

y

3

CJ

CJ

£

CJ

£

CJ

rH

M

rH

U

rH

H

>H

H

Sn

r-H

H

OJ

rH

-J

r-H

1 — i

0)

H

1)

t— 1

■v

4J

UJ

u

a)

1J

■U

1)

4J

CH

3

CO
-J

.5

CO
H

3

13

CO

3

CO

3

M

U

Sh

Sh

Hi

U

:j

u

1)

u

n

-U

M

a)

M

V

4-J

4J

4-1

.a

u

rC

u

C

3

fl

3

cO

c

01

~x>

V

00

cU

HI

00

ai

00

aj

u

H

u

rH

u

CJ

H

u

H

c_>

rH -rH

CO

1

co

SO 0)

g

«J

CO rH

M XI

•H

hJ 3

TJ

3 -H 3 Jh

3 -a a, ai

72

e ©

© #

m ©

© ©

r##i

L# ©i

F© ©1

La ©i
f © el

CX3E3

k o ©J

s

N S cni 4J

I I > «H

> >, >* > 2 cfl

2 O > 2 3 0)

3 co eu 3 H s

2 H > cm -u

° 6 &

5SS

•H ,X1 M-l

i— ( CM CO <r m vO H !N tO <t ^ \C

i_J

X

0

AJ

3

=

3

3

a

■H

U-,

•H

U

H

M

a

W

P5

V

■m

(U

"i

V

c

^

m

■D

en

';1

u

3

3

X)

■H

ai

•H

•H

CJ

a

J

4-J

•H

—1

u

to

w

CXJ

c

4-J

c

fl

-

M

01

c

X

01

•J

3

X

71

01

(30

■H

0)

sz

CM

>

Eg

0

i—

-J
0

■J

1
>

5:

0

i=

c

0)

:~.

\ 3

^

0

u

U

1-

M

pq

-o

-J

—

I*

i

1

1

i

oj

•H

is

_c

■JO

0

m

pa

U

a

a

l—l TJ *W U

a fq §

& fe

O 2

c/3 3

> OM

2 I

u r^ B Tl c

QJ 0 6 N 0)

05 CO H a o

a. h n n ^- in \o i-i om <-o <r m v/o ■— i cm <~"> o- m *x>

74

o ©

© o

o o

75

Antisera to inclusions induced by WMV-1 (Jordan) , WMV-2 (Florida)
and WMV-M each gave strong homologous reactions and none reacted with
their respective purified viruses (at antigen concentrations of 50-100
yg/ml) in SDS double immunodiffusion tests (Figure 16) . The WMV-1 and
WMV-2 inclusion antisera reacted homologously only, whereas the WMV-M
inclusion antiserum reacted heterologously with purified inclusion
preparations (400 lig/ml) or crude extract preparations of either WMV-1
or WMV-2 (Figure 17, Table 4) and with sap containing DMV or PRSV
(Figure 17) . Intragel absorption of WMV-M inclusion antisera with
purified inclusion preparations at 1 mg/ml of WMV-1, WMV-2, or WMV-M
was performed. Intragel absorption wtih WMV-1 or WMV-2 inclusions
resulted in the formation of a precipitin band between WMV-M inclusions
and its homologous antiserum but not between WMV-M inclusion antiserum
and the cross absorbing heterologous antigens. Intragel absorption by
WMV-M inclusions prevented the formation of any precipitin bands by
WMV-1, WMV-2, or WMV-M inclusions and WMV-M inclusion antisera (Figure
18). Intragel absorption of WMV-1 Jordan inclusion antisera with
either WMV-1 Florida and WMV-1 Jordan purified inclusion preparations
(1 mg/ml) was complete (Figure 19).

Enzyme-Linked Immunosorbent Assay (ELISA)

Reciprocal tests were conducted with enzyme labeled gamma globulins
specific for WMV-M, WMV-1 (Florida), and WMV-2 (Florida) and their
corresponding antigens (Table 5). The serum conjugates also were
tested against samples containing the following viruses: WMV-1
Jordan, zucchini yellow fleck virus (kindly supplied by C. Volvas) ,
turnip mosaic virus, potato virus Y, blackeye cowpea mosaic virus,

Figure 16. Reciprocal SDS double immunodiffusion serology of
WMV cylindrical inclusions.

Center wells were charged with the following anitsera:

Jis = WMV-1 Jordan
2is = WMV-2
Ms = WMV-4

Peripheral wells contained the following antigens:

1 = WMV-1 Florida in sap

Ji = WMV-1 Jordan purified inclusions (400 yg/ml)
Jv = WMV-1 Jordan purified virus (50 Ug/ml)

2 = WMV-2 in sap

2i = WMV-2 purified inclusions (400 yg/ml)
2v = WMV-2 purified virus (50 yg/ml)

M = WMV-M in sap
Mi = WMV-M purified inclusions (400 yg/ml)
Mv = WMV-M purified virus (50 yg/ml)

H = Healthy pumpkin sap

Arrow denotes spur formation. Medium consisted of 0.8%
Noble agar, 0.5% SDS, and 1% sodium azide.

77

■^^■■■H

i^Bflk-^HHH

5

•H T3

0)

•H

-o 0

7i

4J

g

N

CU 0]

•H

1

3

6

■u

cfl

>

GO r—

5

cm

.7)

C •

a

-1

•H O

c

•H

4J

>i

o

~*

rH

C3

O T3

■H

0

CO

rH n

X

H

O. cfl

2

i— 1

51

CM ro <r uo vO

c/I

•H

^>

T3

3

C

0
•H

O

4h

■f)

fiv°

0

3

rH

>, y-i o

> o m ^.

•h p a u

E

W rH

CO

^

g

?3

>

ij

3 £>

1

>

1

rH

0 3

u

-H

rH

>

>

~

>

c )

Cfl

M O

cfl

11

Cfl

;> \

O

2

rH

C1J

O -3

CO

3

'2

d

P3

m

O W

ij

w a

0!

U

a

rH

cs

c-i

v-r

m

*.0

0) W

4-1

H
,£3

c

■H

0) >,

'J

'.J

CD

79

U-l

CO

0

d

S)

CO

T3

c

4-1

•H

0

0

rJ

■H

0

en

SO i-l

2

c

_c

H

•H

a

CJ

u

3

•/I

a

—i

•H

00 ••
C 03
■H H
3 H

o a;

-«. B
so\

a. so

CO E T3
•H C

(DOS E

0 iH 30 SO

•H jj E S E E

co C 3 3

3 CO r-! S-l rH Ui

r-H ^ J) ^ J)

0 C (0 CO

C O CO —i CO -H E

•H — i C -U 3 -U 3

co o d o s i-i

-a 3 T-t co -h co ai

01 |H CO CO CO

•H CJ 3 C 3 C -H

'■H C H O rH 0 4-1

•H -H CJ -H CJ -H C

sot;

C3 TJ

£

x

0

3

O

•r-

•H

3

CO

— i

3

a

c

•J

rH CO CU U-i -H

c

c

cfl -

h a 5-s u

a, i cm ai

n > ■ 3

0 z o

co 3 co

.3 « H

03 T3 U rH

C CO <U

rH CO so 3

oj co

60 » r-l

co cn a) <u

S-i I H U

W > .O 3

a Z o <u

h g z a

co Z OJ 3 OJ 3 -H
~ 3 iH -H -H ti co

0 J->

r-t -H

r-H M

1 1)

I

<

4-1

•H Z

U I

3 >

5 tj

ii

a-H-a 3 -a s a, -r-l a.

C04-I-H cfl-H co M aj E

CO -H l-l "3) U T3 CO -H 3

U O U O ^ MO.

33r-HCr-l033

•H ah Th ^ -H a. >-,

i— IS ZZrHr— H H M !N J
tO tU J J J I I I I I rH

0J3J ZZZZZZZZO)

3 CO 3

0) CO
3 -H

81

© m

o

©(© ©

C U T3 -C

g

0)

0

3

3

C

rH

G

TJ

0

_c

X

u

CJ

CJ

en S c s e

C 3 O 3 3
O W -H U Sh

H

n

:r

a)

<U

en

en

j

en

71

3

■H

tH

•H

■H

H

HJ

U

4J

.u

a

3

3

3

3

c

3

•H

CTJ

3

T3 O 0) O O

•H

en

tH

W

en

0)

3.

•H

en

4-1

Oh

U— 1

3

■H

3

3

60

3

IM

•/i

•H

3

■H

i—l

W

—i

H

M

cn

•H

U

tn

H

CJ

3

U

CJ

3

W

3

3

13

3

o-

3

3

Xi

3

3

•H

a

3

a

■H

3

•H

■H

CJ

■H

0,

3

ffl

■H

3

3

"3

3

3

cu

3

3

T)

T7

a,

R)

ci!

•H

-v;

3

w

3

3

•H

•H

S=

"3

"3

H

"3

r-j

CU

-3

T3

w

M

~

M

l-i

0

M

H

r?

U

H

0

3

P.

0

a

H

0

0

0

3

r-i

r-i

^

—

1^4

^

c-1

en

rH

*~i

I-)

•^

|X<

>.

i

1

H
1

H

1

H

1

1)

rH

I

H
I

H
1

rH
1

aJ

H

§j

>

2

>

6

3

|

>

>

3

cu

3

3

3

2

3

rH

2

3

3

3

33

00 2

2

CI

3 3

r-

W

U

(K'

1

■H

4-1 T3
C 3

30

i

u

— I cfl

d

3

3

•n hh MPS

83

© © e
© ©

34

Table 5. ELISA serology of WMV isolates.

ELISA Test Antigens in Sap3 WMV-M WMV-l(Fl)bd WMV-2(Fl)Cd

1. Healthy N. benthamzana

2. WMV-1 Jordan

3. WMV-1 Florida

4. WMV- 2 Florida

5. Zucchini Yellow Fleck Virus

6. WMV-M

7. Healthy Cowpea

8. Turnip Mosaic Virus

9. Potato Virus Y

10. Healthy Bean

11. Healthy Pumpkin

12. PBS-Tween

13. Soybean Mosaic Virus

14. Bean Common Mosaic Virus

15. Healthy Niaotiana edwardsonH

Used at dilutions (w/v) of 10% and 1% in each.

Coating antiserum cone: 1 yg/ml; antiserum- enzyme conjugate: 1/800
dilution,
c
Coating antiserum cone: 1 yg/ml; antiserum-enzyme conjugate: 1/300
dilution.

At dilution of test antigens, value represents the average 0D,ns for
four replications of each test antigen.

0.00

0.00

0.00

0.16

2.40

0.00

0.10

0.88

0.00

0.00

0.01

5.20

0.00

0.04

0.00

1.10

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

3.40

0.00

0.00

0.20

0.00

0.00

0.00

85

bean common mosaic virus (Siratro isolate; Lima, 1978), and healthy
saps of pumpkins, bountiful bean, cowpea, Nicotiana benthamiana and
N. edwardsonii Christie and Hall. Only the homologous reactions with
the Moroccan conjugate was clearly positive, though a weak reaction at
the threshold level (Rochow and Carmichael, 1979) occurred with the
Florida and Jordan isolates of WMV-1. The WMV-1 (Florida) conjugate
reacted only with the Florida and Jordan isolates of WMV-1. The Florida
WMV-2 conjugate gave a strong homologous reaction, and heterologous
reactions with sovbean mosaic virus and bean common mosaic virus.

DISCUSSION

This study provides new evidence that the relationships among
watermelon mosaic viruses are considerably more complex than was realized
previously (Milne and Grogan, 1969; van Regenmortel, 1977). The evidence
provided herein, however, supports the observations of Webb and Scott
(1965) and Purcifull and Hiebert (1979) that WMV-1 and WMV-2 are
serologically distinct.

Evidence also is presented in this dissertation that WMV-M is indeed
serologically distinct from WMV-1 and WMV-2, as suggested previously by
Purcifull and Hiebert (1979). Gel immunodiffusion tests with antisera
specific for the capsid proteins of WMV-1, WMV-2, or WMV-M, showed that
the three isolates are serologically distinct. These distinctions
applied to antisera collected up to 4 months after the initial injections;
one of the WMV-M antisera collected later than 4 months after immunization
reacted with WMV-1. These results were generally upheld by the ELISA
tests. The WMV-M antiserum showed a weak heterologous reaction with
WMV-1, but failed to react either with WMV-2 or the zucchini yellow
fleck virus from Italy in ELISA tests. Neither the WMV-1 nor WMV-2
antisera reacted with WMV-M in ELISA.

It is of limited value to define WMV-1 as isolates of WMV which
have a host range confined to the Cucurbitaceae (Webb and Scott, 1965).
Purcifull and Hiebert (1979) obtained serological reactions of identity
using Florida WMV-1 antiserum and Mediterranean isolates of WMV whose
host ranges included members of the Chenopodiaceae as well as the

S6

87

Cucurbitaceae. In this study, antisera were made to the virus and the
cylindrical inclusions of one of the Mediterranean isolates, WMV-1
Jordan. Intragel absorption in reciprocal double immunodiffusion tests
failed to detect differences between the capsids of the two viruses.
In addition, these two WMV-1 isolates produced similar capsid digest
patterns following CNBr cleavage. Intragel cross-absorption of WMV-1
Jordan inclusion antisera with WMV-1 Florida purified inclusions was
complete. These tests confirmed the close relationship between WMV-1
Florida, which has no known hosts outside the Cucurbitaceae (Purcifull
and Hiebert, 1979), and the Jordan isolate, which has a broader host
range (Martelli and Russo, 1976). The concept (Webb and Scott, 1965)
that North American isolates of WMV-1 are limited to the Cucurbitaceae
may also need revision, because an isolate of WMV has recently been
found in South Carolina which caused local lesions on C. amaraniieolor
but which was closely related serologically to WMV-1 Florida (personal
communication by 0. W. Barnett) .

The serological tests indicated that WMV-1, WMV-2, and WMV-M
inclusions were distinct, although the Moroccan isolate was related
to both WMV-1 and WMV-2. Antisera to inclusions gave stronger reactions
in SDS double immunodiffusion tests with purified inclusions than with
sap extracts. The medium of Tolin and Roane (1975) gave stronger
reactions, but was more likely to result in nonspecific reactions, than
the medium consisting of 0.8% agar, 0.5% SDS, and 1.0% sodium azide
(Purcifull and Batchelor, 1977). These results suggest that it could
be useful to investigate the efficacy of these media, in order to
optimize results in serological detection of inclusion body proteins.

The WMV-1, WMV-2, and WMV-M isolates were compared in other ways.
Unlike WMV-1 and WMV-2, the WMV-M was unstable in cesium chloride. Both
WMV-i and WMV-M were unstable in n-butanol, although this solvent was
useful for WMV-2 purification. The CNBr derived peptide pattern of WMV-M
capsid proteins had a degree of similarity with the peptide protein of
WMV-1 isolates when electrophoresed at pH 4.5. However, the peptide
pattern of WMV-M was distinct from those of the WMV-1 isolates when the
pH of the electrode and separating gel buffers were lowered approximately
one unit. The WMV-2 peptide pattern was always distinct from WMV-M and
both WMV-1 isolates. Molecular weights of the cylindrical inclusion
proteins of WMV-1 (both the Florida and Jordan isolates), WMV-2, and WMV-M
all averaged 69,000 daltons. The peptide fragment patterns following
digestion of inclusions with S. aureus protease indicated that WMV-M
patterns were distinct from those of WMV-1 or WMV-2.

This dissertation points to the need for augmenting the standard
techniques presently used to determine strain relationships among poty-
viruses. Peptide mapping of virus-specified proteins is one such approach.
The recently evolved techniques for immunochemical analysis of CNBr cleaved
fragments (Doyen and Lapresle, 1979; Vita et at., 1979) may in the future
be used to map antigenic sites and to expose new sites which may aid
further in determining serological relationships.

In conclusion, this study supports the proposition that there are
at least three serologically distinct viruses involved in the WMV complex.
The three types are represented by the WMV-1, WMV-2, and WMV-M isolates.
It would be of particular interest to determine the serological relation-
ship of WMV-M to South African isolates (van Regenmortel et al., 1962),
which reportedly also may differ from WMV-1 and WMV-2 (.<Iebb and Scott, 1965).

LITERATURE CITED

Adlerz, W. C. 1969. Distribution of watermelon mosaic virus 1 and 2 in
Florida. Proc. Fla. State Horticultural Soc. 81: 161-165.

Anderson, C. W. 1951a. Further observations on some Cucurbit viruses
from central Florida. Plant Dis. Rep. 35: 396-398.

Anderson, C. W. 1951b. Viruses of Cucurbits in central Florida.
Proc. Fla. State Horticultural Soc. 64: 109-112.

Anderson, C. W. 1954. Two watermelon mosaic virus strains from central
Florida. Phytopathology 44: 198-202.

Arteaga, M. P. L. , J. B. Quiot, and J. P. Leroux. 1976. Mise en

evidence d'une souche de watermelon mosaic virus (WMV-II) dans
sud-est de la France. Ann. Phytopathol. 8: 347-353.

Auger, J. G. , 0. Escaffi, and F. S. Nome. 1974. Occurrence of water-
melon mosaic virus 2 on Cucurbits in Chile. Plant Dis. Rep. 58:
599-602.

Bakker, W. 1971. Notes on East African plant virus diseases: II.

Courgette leaf distortion incited by watermelon mosaic virus. East
African Agric. and Forestry J. 82: 78-85.

Ball, E. M. 1974. Serological tests for the identification of plant
viruses. Am. Phytopathol. Soc, St. Paul, Minn.: 31 pp.

Batchelor, D. L. 1974. Immunogenicity of sodium dodecyl sulfate

denatured plant viral inclusions. Ph.D. dissertation, University
of Florida, Gainesville. 81 pp.

Baum, R. H. , D. E. Purcifull, and E. Hiebert. 1979. Purification and
serology of a Moroccan isolate of watermelon mosaic virus (WMV) .
Phytopathology 69: 1021-1022.

Bhargava, B. 1977. Some properties of two strains of watermelon mosaic
virus. Phytopathol. Z. 88: 199-208.

Brandes, J. 1964. Identif izierung von gestreckten pf lanzenpathogenen
Viren auf morphologischer Grundlage. Mitt. Biol. Bundesanst.
Land Forstwirtsch. (Berlin-Dahlem) . 110: 1-130.

Campbell, R. N. 1971. Squash mosaic virus. CM. I . /A. A. B. Descriptions
of Plant Viruses No. 43. 4 pp.

89

90

Christie, R. G. 1967. Rapid staining procedures for differentiating

plant virus inclusions in epidermal strips. Virology 31: 268-271.

Christie, R. G. and J. R. Edwardson. 1977. Light and electron microscopy
of plant virus inclusions. Fla. Agric. Exp. Stn. Monogr. Ser. 9.
150 pp.

Christie, S. R. and W. E. Crawford. 1978. Plant virus range of
Niaotiana benthcartiana. Plant Dis. Rep. 62: 20-22.

Clark, M. F. and A. N. Adams. 1977. Characteristics of the microplate
method of enzyme-linked immunosorbent assay for the detection of
plant viruses. J. Gen. Virol. 34: 475-483.

Cleveland, D. W. , S. G. Fischer, W. M. Kirschner, and U. K. Laemmli.
1977. Peptide mapping by limited proteolysis in sodium dodecyl
sulfate and analysis by gel electrophoresis. J. Biol. Chem.
252: 1102-1106.

Cohen, S. and F. E. Nitzany. 1963. Identity of viruses affecting
Cucurbits in Israel. Phytopathology 53: 193-196.

Damirdagh, I. S. and R. J. Shepherd. 1970. Purification of the

tobacco etch and other viruses of the potato Y group. Phyto-
pathology 60: 132-142.

Demski, J. W. 1968. Local lesion reactions of Chenopodium species to
watermelon mosaic virus 2. Phytopathology 58: 1196-1197.

Demski, J. W. and J. H. Chalkley. 1974. Influence of watermelon
mosaic virus on watermelon. Plant Dis. Rep. 58: 195-198.

Dougherty, W. G. 1979. Translation of poty virus RNA in a rabbit

reticulocyte lysate. Ph.D. dissertation, University of Florida,
Gainesville. 91 pp.

Doyen, N. and C. Lapresle. 1979. Properties immunochimiques des

fragments obtenus par degradation de la serum albumine humaine
avec le bromure de cyanogene. Ann. Immunol. (Inst. Pasteur)
130: 22-334.

Ebrahim-Nesbat , F. 1974. Distribution of watermelon mosaic viruses 1
and 2 in Iran. Phytopathol. Z. 79: 352-358.

Edwardson, J. R. 1966. Electron microscopy of cytoplasmic inclusions
in cells infected with rod shaped viruses. Am. J. Bot. 53:
359-364.

Edwardson, J. R. 1974a. Some properties of the potato Y-group. Fla.
Agric. Exp. Stn. Monogr. Ser. No. 4. 398 pp.

Edwardson, J. R. 1974b. Host ranges of viruses in the PVY-group.
Fla. Agric. Exp. Stn. Monogr. Ser. No. 5.

91

Edwardson, J. R. , D. E. Purcifull, and R. G. Christie. 1968. Structure
of cytoplasmic inclusions in plants infected with rod-shaped
viruses. Virology 34: 250-263.

Fischer, H. U. and B. E. L. Lockhart. 1974. Serious losses in Cucurbits
caused by watermelon mosaic virus in Morocco. Plant Dis. Rep.
58: 143-146.

Gates, L. F. and J. F. Bronskill. 1976. Occurrence of watermelon mosaic
virus in glasshouse cucumbers in southwestern Ontario. Plant Dis.
Rep. 60: 964-966.

Ghosh, S. K. and S. Mukhopadhyay . 1979. Viruses of pumpkin (Cuaurbita
moschata) in west Bengal. Phytopathol. Z. 94: 172-184.

Gonzaiez-Garza, R. , D. J. Gumpf, A. N. Kishaba, and G. W. Bohn. 1979.
Identification, seed transmission, and host range pathogenicity
of a California isolate of melon necrotic spot virus. Phyto-
pathology 69: 340-345.

Greber, R. S. 1969. Viruses infecting Cucurbits in Queensland.
Queensland J. Agric. and Anim. Sci. 26: 145-171.

Greber, R. S. 1978. Watermelon mosaic virus 1 and 2 in Queensland
Cucurbit crops. Aust. J. Agric. Res. 29: 1235-1245.

Grogan, R. G. , D. H. Hall, and K. A. Kimble. 1959. Cucurbit mosaic
viruses in California. Phytopathology 49: 366-376.

Gross, E. 1967. The cyanogen bromide reaction. Chapt. 27: 238-255,
in Methods in Enzymology, Vol. XI, Academic Press, New York.

Gross, E. and B. Witkop. 1962. Nonenzymatic cleavage of peptide bands:
The methionine residues in bovine pancreatic ribonuclease. J.
Biol. Chem. 237: 1856-1860.

Harrison, B. D. , J. T. Finch, A. J. Gibbs, M. Hollings, R. J. Shepherd,
V. Valenta, and C. Wetter. 1971. Sixteen groups of plant
viruses. Virology 45: 356-363.

Hebert, T. T. 1963. Precipitation of plant viruses by polyethylene
glycol. Phytopathology 53: 362.

Hein, A. 1977. Uber ein Auftreten des Wassermelonenmosaik Virus 1 an

Zucchini (Cucurbita pevo L. var. givcmontiiria Alef.)in Suddentschland
Phytopathol.. Z. 89: 221-228.

Hiebert, E. and J. F. McDonald. 1973. Characterization of some proteins
associated with viruses in the potato Y group. Virology 56: 349-
361.

92

Hiebert, E. and J. G. McDonald. 1976. Capsid protein heterogeneity
in turnip mosaic virus. Virology 70: 144-150.

Hiebert, E. , D. E. Purcifull, R. G. Christie, and S. R. Christie. 1971.
Partial purification of inclusions induced by tobacco etch virus
and potato virus Y. Virology 43: 638-646.

Hiebert, E. , J. H. Tremaine, and B. Ronald. 1979. Characterization of
the capsid protein of tobacco etch virus before and after in situ
degradation. Phytopathology 69: 1031.

Hitchborn, J. H. and G. J. Hills. 1965. The use of negative staining
in electron microscopic examination of plant viruses in crude
extracts. Virology 27: 528-450.

Horvath, J. , N. Juretic, W. H. Besada, and G. Kuroli. 1975. Two viruses
isolated from Patisson (Cucurbita pepo L. var. patissonina Greb.
f. radiata Nois.) a new vegetable natural host in Hungary. I,. Water-
melon mosaic virus (general). Acta Phytopathol. Acad. Scient.
Hung. 10: 93-111.

Inouye, T. 1964. A virus disease of pea caused by watermelon mosaic
virus. Berichte des Ohara Instituts fur Landwirtschaf tliche
Biologie, Okayama Universitie. 12: 133-143.

Koenig, R. 1978. ELISA in the study of homologous and heterologous
reactions in plant viruses. J. Gen. Virol. 40: 309-318.

Komm, D. A. and G. N. Agrios. 1978. Incidence and epidemiology of
viruses affecting Cucurbit crops in Massachusetts. Plant Dis.
Rep. 62: 746-750.

Laemmli, U. K. 1970. Cleavage of structural proteins during the

assembly of the head of bacteriophage T4. Nature 227: 680-685.

Lastra, R. 1968. Occurrence of Cucurbit viruses in Venezuela. Plant
Dis. Rep. 52: 171-174.

Lima, J. A. A. 1978. Blackeye cowpea mosaic virus: Purification,

partial characterization, serology, and immunochemical and cyto-
logical techniques for detection of virus-infected legume seeds.
Ph.D. dissertation, University of Florida, Gainesville. 154 pp.

Lima, J. A. A., D. E. Purcifull, and E. Hiebert. 1979. Purification,
partial characterization, and serology of blackeye cowpea mosaic
virus. Phytopathology 69: 1252-1258.

Lindberg, G. D. , D. H. Hall, and J. C. Walker. 1956. A study of melon
and squash mosaic virus. Phytopathology 46: 489-495.

Martelli, G. P. and M. Russo. 1976. Unusual cytoplasmic inclusions
induced by watermelon mosaic virus. Virology 72: 352-362.

93

Mauer, H. R. 1971. Disc electrophoresis and related techniques of
polyacrylamide gel electrophoresis. Second Edition, Walter de
Gruyter, Berlin-New York.

McDonald, J. G. and E. Hiebert. 1975. Characterization of the capsid
and cylindrical inclusion proteins of three strains of turnip
mosaic virus. Virology 63: 295-303.

Milbrath, G. M. and M. R. Nelson. 1968. Watermelon mosaic virus 2 and

Cucurbit latent virus: A new evaluation. Phytopathology 58: 687-689.

Milne, K. S. and R. G. Grogan. 1969. Characterization of watermelon
mosaic virus strains by serology and other properties. Phyto-
pathology 59: 809-818.

Milne, K. S., R. G. Grogan, and K. A. Kimble. 1969. Identification of
viruses infecting Cucurbits in California. Phytopathology 59:
819-828.

Molnar, A. and K. Schmelzer. 1964. Beitrage zur Kenntnis des Wasser-
melonenmosaik Virus. Phytopathol. Z. 51: 361-384.

Nelson, M. R. and D. M. Tuttle. 1969. The epidemiology of cucumber

mosaic and watermelon mosaic 2 of cantaloups in an arid climate.
Phytopathology 59: 849-856.

Ouchterlony, 0. 1962. Dif fusion- in-gel methods for immunological
analysis II. Prog. Allergy 6: 30-154.

Prowidenti, R. and W. T. Schroeder. 1970. Epiphytotic of watermelon
mosaic among Cucurbitaceae in central New York in 1969. Plant
Dis. Rep. 54: 744-748.

Purcifull, D. E. 1966. Some properties of tobacco etch virus and its
alkaline degradation products. Virology 1: 8-14.

Purcifull, D. E. and D. L. Batchelor. 1977. Immunodiffusion tests with
sodium dodecyl sulfate (SDS)-treated plant viruses and plant viral
inclusions. Fla. Agric. Exp. Stn. Bull. (Tech.) 788. 39 pp.

Purcifull, D. E. and J. R. Edwardson. 1967. Watermelon mosaic virus:
Tubular inclusions in pumpkin leaves and aggregates in leaf
extracts. Virology 32: 393-401.

Purcifull, D. E. and E. Hiebert. 1979. Serological distinction of
watermelon mosaic virus isolates. Phytopathology 69: 112-116.

Purcifull, D. E. , E. Hiebert, and J. G. McDonald. 1973. Immunochemical
specificity of cytoplasmic inclusions induced by viruses in the
potato Y group. Virology 55: 275-279.

94

Purcifull, D. E. and T. A. Zitter. 1973. A serological test for distin-
guishing bidens mottle and lettuce mosaic viruses. Proc. Fla. State
Hort. Soc. 86: 143-145.

Quiot, J. B. , F. Kaan, and M. Beramis. 1971. Identification d'une souche
de la mosaique de la pasteque (watermelon mosaic virus 1) aux
Antilles Francaises. Ann. Phytopathol. 3: 125-130.

Raggozzino, A. and D. Stefanis. 1977. Neoformazioni cupoliformi su zucchine
infette dal virus del mosaico del cocomero (watermelon mosaic virus
1). Annali della Fac. Sci. Agr. , Univ. Napoli in Portici Ser IV 11:
33-41.

Rochow, W. F. and L. E. Carmichael. 1979. Specificity among barley

yellow dwarf viruses in enzyme linked immunosorbent assay. Virology
95: 415-420.

Schmelzer, K. 1969. Some more recent knowledge of Cucurbit viruses.

Proc. 6th Conf. Czechoslovak Plant Virologists, Olomouc. 1967. Acad.
Publ. House, Czechoslovak Acad. Sci., Prague, pp. 208-210.

Schmelzer, K. and D. Milicic. 1966. Zur Kenntnis der Verbreitung des

Wassermelonenmosaik Virus in Europa und seiner Fahigkeit zur Bildung
von Zelleinschlusskorpern. Phytopathol. Z. 57: 8-16.

Shanmugasundaram, S., M. Ishii, J. C. Gilbert, and H. Nagai. 1969.
Cucurbit virus studies in Hawaii. Plant Dis. Rep. 53: 70-74.

Shawkat, A. L. B., and G. I. Fegla. 1979. Identification of two viruses from
eggplant and Cucurbita. peyo in Iraq. Plant Dis. Rep. 63: 235-238.

Shepard, J. F. 1968. Electron microscopy of subtilisin-treated tobacco
etch virus nuclear and cytoplasmic inclusions. Virology 36: 20-29.

Shepard, J. F. and T. A. Shalla. 1969. Tobacco etch virus cylindrical
inclusions: Antigenicity unrelated to the causal virus. Virology
38: 185-188.

Stakic, D. and V. Nikolic. 1966. Watermelon mosaic, a new disease in
Yugoslavia. Contemporary Agric. 3: 289-302.

Studier, F. W. 1973. Analysis of bacteriophage T7 early RNAs and proteins
on slab gels. J. Mol. Biol. 79: 237-248.

Thomas, W. 1971a. Watermelon mosaic virus, a disease of Cucurbits in
New Zealand. New Zealand J. Agric. Res. 14: 235-241.

Thomas, W. 1971b. The incidence and economic importance of watermelon
mosaic virus. New Zealand J. Agric. Res. 14: 242-247.

Toba, H. H. 1962. Studies on the host range of watermelon mosaic virus
in Hawaii. Plant Dis. Rep. 46: 409-410.

Tolin, S. A. and C. W. Roane. 1975. Identification and distribution of soybean
viruses in Virginia. Proc. Amer. Phytopath. Soc. 2: 129 (PO-40) .

95

van Regenmortel, M. H. V. 1960. Zone electrophoresis and electron
microscopy of a watermelon mosaic virus from South Africa.
Virology 12: 127-130.

van Regenmortel, M. H. V. 1961. Purification of a watermelon mosaic
virus. South African J. Agric. Sci. 4: 405-414.

van Regenmortel, M. H. V. 1971. Watermelon mosaic virus. C.M.L /A. A.B.
Description of Plant Viruses No. 63. 4 pp.

van Regenmortel, M. H. V. 1977. Watermelon mosaic virus. In Insect
and Plant Viruses: An Atlas, (K. Maramorosch, ed.), pp. 323-329.
Academic Press, New York.

van Regenmortel, M. H. V. , J. Brandes, and R. Bercks. 1962. Investi-
gations on the properties of watermelon mosaic virus. Phytopathol.
Z. 45: 205-216.

Vita, C. , A. Fontana, J. R. Seeman, I. M. Chaiken. 1979. Conformational
and immunochemical analysis of the cyanogen bromide fragments of
thermolysin. Biochemistry 18: 3023-3031.

Webb, R. E. 1963. Local-lesion hosts for some isolates of watermelon
mosaic virus. Plant Dis. Rep. 47: 1036-1038.

Webb, R. E. 1965. Luff a aautangula for separation and maintenance of
watermelon mosaic virus 1 free of watermelon mosaic virus 2.
Phytopathology 55: 1379-1380.

Webb, R. E. 1971. Watermelon mosaic viruses 1 and 2 in squash on the
Atlantic seaboard. Plant Dis. Rep. 55: 132-135.

Webb, R. E. and G. W. Bohn. 1961. A virus latent in some cucurbits.
Plant Dis. Rep. 45: 677-679.

Webb, R. E. and H. A. Scott. 1965. Isolation and identification of
watermelon mosaic virus 1 and 2. Phytopathology 55: 895-900.

Weber, K. and M. Osborn. 1969. The reliability of molecular weight
determinations by dodecyl sulfate - polyacrylamide gel electro-
phoresis. J. Biol. Chem. 244: 4406-4412.

Zurawski, D. B. 1979. Some biological and serological properties of

bidens mottle virus isolated from Fittonia. M.S. thesis, Univer-
sity of Florida, Gainesville. 91 pp.

BIOGRAPHICAL SKETCH

Robert H. Baum was born in Lynbrook, New York, on April 28, 1942,
and grew up in South Carolina, New York, and Florida. He graduated
with an A.B. degree from Goddard College in 1966 with majors in chemistry
and biology. Following graduation, he worked in engineering positions
at Cape Kennedy, Florida, and attended the University of Florida's
Graduate Engineering Extension System on a part-time basis. After
three years, he transferred to the Kwajalein Missile Range in the
Marshall Islands, Central Pacific, where he worked on the Spartan
Antiballistic Missile program. In 1972, he became a graduate student
in the Botany Department where he held a teaching assistantship. In
1974, he received his M.S. degree under the supervision of Dr. H. C.
Aldrich. He transferred to the Plant Pathology Department in 1975,
where he was awarded a research assistantship under the supervision of
Dr. Dan Purcifull. He is presently a Visiting Assistant Professor in
the Plant Pathology Department at Clemson University. He is married
to the former Diane Mundhenk and they are looking forward to the birth
of their first child.

96

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.

6%U^. £ . //C'tA

Dan E. Purcifull, Qfaairman
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.

0- 7 <<

,'J . R. Edwardson
■^Professor of Agronomy

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.

& , / / '/"

^A

s.'~ Lr

E. Hiebert

Associate 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.

D. R. Pring

Associate 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.

AG

./jcgWa

D. A. Roberts

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.

R. E. Stall

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. ^

U-J.

I. K. Vasil

Graduate Research Professor of Botany

This dissertation was submitted to the Graduate Faculty of the College of
Agriculture and to the Graduate Council, and was accepted as partial
fulfillment of the requirements for the degree of Doctor of Philosophy.

March, 1980 /\

Dean,/(/ollege of Agriculture

Dean, Graduate School