Serological and molecular comparisons of coronaviruses
by Susan Herbert Goss
A thesis submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in
Microbiology
Montana State University
© Copyright by Susan Herbert Goss (1984)
Abstract:
The antigenic interrelationships among viruses of the family Coronaviridae were examined. Four
mammalian coronaviruses from two different antigenic groups were compared using serological and
molecular techniques. The viruses selected were murine hepatitis virus A59 (A59V), bovine
coronavirus (BCV), canine coronavirus (CCV) and transmissible gastroenteritis virus (TGEV).
Hyperimmune ascitic fluids were prepared in mice to provide a homogeneous source of antibody.
Virus-infected cell lysates were used as immunogens to generate antibodies against intracellular as well
as structural proteins. Antibodies directed against cellular proteins were removed by absorption with
uninfected cells. The ascitic fluids compared favorably to hyper immune serum in immunofluorescence
and virus-neutralization tests. The intracellular, proteins of A59V, BCV, CCV and TGEV were
identified by immunoprecipitation of infected-cell lysates with hyperimmune ascitic fluids. Reciprocal
cross-immunofluorescence and neutralization tests showed two-way cross-reactions between A59V and
BCV, and between CCV and TGEV. Immunoprecipitation of virus-infected cell lysates showed that the
serological relatedness was based on common determinants on each of the major viral antigens: the
peplomer, nucleocapsid and membrane. Monoclonal antibodies (MAbs) against each virus were
isolated using hybridoma technology. The MAbs were tested for virus specificity by
immunofluorescence, polypeptide specificity by immunoprecipitation, and virus neutralization activity
by a plaque reduction test. They were also tested for cross-reaction with the related virus. Most of the
MAbs isolated recognized the nucleocapsid protein, which is the major polypeptide synthesized in
infected cells. MAbs which neutralized virus infactivity were directed against the peplomeric proteins.
The immunoprecipitation and MAb studies suggest that TGEV and CCV are more closely related than
AS9V and BCV. SEROLOGICAL AND MOLECULAR COMPARISONS
OF CORONAVIRUSES
by
Susan Herbert Goss
A thesis submitted in partial fulfillment
of the requirements for the degree
of
i
Doctor of Philosophy
in
Microbiology .
MONTANA STATE UNIVERSITY
Bozeman, Montana
December 1984
APPROVAL
of a thesis submitted by
Susan Herbert Goss
This thesis has been read by each member of the thesis
committee and has been found to be satisfactory regarding
content, English usage, format, citations, bibliographic
style, and consistency, and is ready^ fip fy s u b jffls s z x m fto the
College of Graduate Studies.
,
\viW I
Chairpers
Date
Sj^aduate JSbmmittee
Approved for the Major
jor Department
/s? ^ g?(D Date
ft1/
Head^ Major Department
Approved for the College of Graduate Studies
Date
Graduate Dean
© 1985
SUSAN HERBERT GOSS
All Rights Reserved
iii
STATEMENT OF PERMISSION TO USE
In presenting this thesis in partial fulfillment of
the requirements for a doctoral degree at Montana State
University,
I agree that the Library shall make it
available to borrowers under the rules of the Library.
I
further agree that the copying of this thesis is allowable
only for scholarly purposes, consistent with "fair use" as
prescribed in the U.S. Copyright Law. Requests
for
extensive copying or reproduction of this thesis should be
referred to University Microfilms International,
300 North
Zeeb Road, Ann A r b o r , M i c h i g a n 48106, to whom I have
granted "the exclusive right to reproduce and distribute
copies of the dissertation in and from microfilm and the
right to reproduce and distribute by abstract in any
format."
Signature
Date
/H J
aMj
.
A/
iv
TABLE OF CONTENTS
Page
LIST OF TABLES.......................................
LIST OF FIGURES..............................
vi
vii
ABSTRACT.... .........................,...............
viii
INTRODUCTION.........................................
I
Properties of Coronaviruses....................
Antigenic Interrelationships.........^ .........
Molecular Immunology............................
Murine Hepatitis Virus AS9 ......................
Bovine Coronavirus..............................
Transmissible Gastroenteritis Vir u s............
Canine Coronavirus..............................
Methods for Investigating Antigenic
Relationships...............
Experimental Design........^ ....................
2
4
8
10
12
14
15
MATERIALS AND METHODS................................
20
Chemicals and Med i a ..............................
Virus Strains and Cell Lines....................
Preparation of Cell Lysates.....................
Plaque Assay.....................................
Immunization of M i c e ............................
Harvesting of Serum and Ascitic Fluid..........
Absorption of Serum and Ascitic Fluid..........
Preparation of Monoclonal Antibodies...........
Immunofluorescence Assay........................
Plaque Reduction Test...........................
Radiolabeling of Intracellular Proteins.......
Immunoprecipitation of Virus-Specific Proteins.
SDS-Polyacrylamide Gel Electrophoresis.........
20
21
22
22
' 23
24
24
25
26
27
28
29
30
16
18
V
TABLE OF CONTENTS (continued)
Page
RESULTS........... ............... ...................
31
Establishment of Virus-Cell Systems...........
Preparation and Testing of Anti-Viral
Ascitic Fluids............................
Intracellular Virus-Specific Proteins.........
Serological Interrelationships ................
Antigenic Analysis using Monoclonal
Antibodies.................................
33
37
41
DISCUSSION.............. .............................
55
Serological Comparisons........................
Molecular Comparisons ..........................
Intracellular Virus-Specific Proteins.........
Monoclonal Antibodies..........................
Conclusions.....................................
57
58
58
60
62
LITERATURE CITED................. ............... . • •
31
49
64
vi
LIST OF TABLES
Table
Page
1.
Members of theCoronaviridae. . .■...................
4
2.
Antigenic cross-reactions among coronaviruses...
5
3.
Virus neutralization titers of hyperimmune
ascitic fluids.....................................
434
Characteristics of monoclonal antibodies...... .
50
4.
vii
LIST OF FIGURES
Figure
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
Time course for coronavirus protein synthesis
in cells infected with A59V',B C V fCCV and TGEV..
L
Comparison of unabsorbed and absorbed
hyperimmune ascitic fluids.....................
Page
34
36
Comparison of virus-neutralizing activity of
anti-A5 9V and BCV serum and ascitic fluid....
38
Comparison of virus-neutralizing activity of
anti-CCV and TGEV serum and ascitic fluid....
39
Intracellular virus-specific proteins of
coronaviruses AS9, B C V , CCV and TGE V ..........
40
Antigenic cross-reactions among coronaviruses
by immunofluorescence..........................
42
Immunoprecipitation of intracellular MHV-A S 9
proteins by homologous and heterologous
ascitic fluids..................................
44
Immunoprecipitation of intracellular BCV
proteins by homologous and heterologous
hyperimmune ascitic fluids.....................
45
Immunoprecipitation of intracellular CCV
proteins by homologous and heterologous
ascitic fluids..................................
46
Immunoprecipitation of intracellular TGEV
proteins by homologous and heterologous
ascitic fluids..................................
47
Analysis of monoclonal antibody specificities
by immunoprecipitation.........................
51
Cross-immunoprecipitation of viral proteins by
monoclonal antibodies.........................
54
viii
ABSTRACT
The antigenic interrelationships among viruses of the
family Coronaviridae were examined.
Four mammalian
coronaviruses from two different antigenic groups were
compared using serological and molecular techniques. The
viruses selected were murine hepatitis virus A59 (A59V),
bovine coronavirus (BCV), canine coronavirus (CCV) and
transmissible gastroenteritis virus (TGEV).
Hyperimmune
ascitic fluids were prepared in mice to provide a
homogeneous source of antibody.
Virus-infected cell
lysates were used as immunogens to generate antibodies
against intracellular as well as structural proteins.
Antibodies directed against cellular proteins were removed
by absorption with uninfected cells.
The ascitic fluids
compared favorably to hyper immune serum in immunofluoresbence and virus-neutralization tests.
The
intracellular, proteins of A59V, BCV, CCV and TGEV were
identified by immunoprecipitation of infected-cel I lysates
with hyperimmune ascitic fluids. Reciprocal cross - '
immunofluorescence and neutralization tests showed two-way
cross-reactions between A59V and BCV, and between CCV and
TGEV.
Immunoprecipitation of virus-infected cell lysates
showed that the serological relatedness was based on
common determinants on each of the major viral antigens:
the peplomer, nucleocapsid and membrane.
Monoclonal
antibodies (MAbs) against each virus were isolated using
hybridoma technology.
The MAbs were tested for virus
specificity by immunofluorescence, polypeptide specificity
by immunoprecipitation, and virus neutralization activity
by a plaque reduction test. They were also tested for
cross-reaction with the related virus.
Most of the MAbs
isolated recognized the nucleocapsid protein, which is the
major polypeptide synthesized in infected cells.
MAbs
which neutralized virus infactivity were directed against
the peplomeric proteins. The immunoprecipitation and MAb
studies suggest that TGEV and CCV are more closely related
than AS9V and BCV.
I
INTRODUCTION
The power of immunological techniques in research and
clinical virology is well established.
The serological
characterization of viruses is useful for diagnosis and
epidemiological studies of viral diseases.
In fact, the
identification of specific antibody is frequently the only
available method of diagnosis, as many pathogenic viruses
are difficult to isolate and characterize in vitro.
A
knowledge of antigenic interrelationships is valuable when
identifying and classifying viral isolates and for
investigation of evolutionary patterns within virus
families.
In spite of recent advances in viral immunology
made possible by molecular cloning and monoclonal antibody
technology, much remains to be done in clarifying the
antigenic relationships within families of human and
animal viruses.
The purpose of my research was to
establish a simple and effective method for studying
antigenic groups of viruses and to use this syst e m to
investigate at the molecular level the immunological
interrelationships among several mammalian coronaviruses.
2
Properties of Coronaviruses
The Coronaviridae are a heterogeneous family of
pathogenic viruses which cause a wide variety of diseases
in many animals including humans.
Coronavirions are
pleomorphic particles 60-220 nanometers in diameter
surrounded by a lipid bilayer envelope studded with clubshaped spikes or peplomers.
It is this "corona" of
surface projections as seen in electron micrographs which
led to the naming of the group by an international
committee in 1968 (72).
The genome is a large
(6-8
megadaI tons) single-stranded RNA molecule of positive
polarity.
Virus replication takes place in the cytoplasm
and virions are released by budding through the membranes
of the endoplasmic reticulum, unlike other RNA viruses
which bud
from the plasma membrane (51,57,68,72).
Virions contain from 3 to 7 structural proteins which
seem to fall into three classes.
A phosphoryI at ed
protein (pp), N , of 50-60 ki Ioda I tons (kd) is located
internal Iy and is associated with the viral genome
(51,57,76,79).
The second family of structural
proteins
is a heterogeneous glycoprotein (gp) species, M , of 20-3 0
kd which often appears as several bands on SDSpolyacry I amide gels (65,66,67). - This protein is mostly
embedded in a lipid bi-layer and forms part of the viral
envelope.
A short glycosylated portion (5 kd) protrudes
3
from the envelope and can be removed by treatment with
hromelain (66,68).
The M protein determines the location
of viral budding and interacts with the nucleocapsid (68).
Antibodies directed against the M protein can neutralize
virus infectivity in the presence of complement (10).
large (125-200 kd) glycoprotein,
virion peplomers (17,66,68).
A
P, is associated with the
Biological activities
associated with this protein include binding of virions to
cell membrane receptors (28,68,75),
neutralizing antibody
(10,27,28).
induction of
(19,24,2 7,55) and cell fusion
Some coronaviruses may have more than one
peplomeric glycoprotein.
Monoclonal antibodies to bovine
coronavirus (BCV) define another, smaller glycoprotein
which is also associated with the virion surface
projections and elicits neutralizing antibody (26,74,75).
This structural protein is probably responsible for the
hemagglutinating activity of BCV
(26,75).
The current members of the Coronaviridae are listed in
Table I.
Several of the coronaviruses were original Iy
placed into the group on the basis of morphology alone.
The list has undergone frequent revision with the improved
sensitivity and specificity of available serological and
molecular techniques for the analysis of coronaviruses.
4
Table I. Members of the Coronaviridae
Natura I
Host
Virus
Avian infectious bronchitis
virus (IBV)
Bluecomb disease virus (TCV)
Bovine coronavirus (BCV)
Canine coronavirus (CCV)
Feline enteric coronavirus (FECV)
Feline infectious peritonitis
virus (FIPV)
Hemagglutinating encephalo­
myelitis virus (HEV)
Human coronavirus (HCV)
Human enteric coronavirus (HECV)
Isolates SD and SK
Murine hepatitis virus (MHV)
Parrot Coronavirus (PCV)
Pleural effusion disease
virus (RbCV)
Porcine virus C M - I l l
Rabbit enteric coronavirus
(RbECV)
Rat coronavirus (RCV)
Sialodacryoadenitis virus .(SDAV)
Transmissible gastroenteritis
virus (TGEV)
Predominant
Type of disease
Chicken
Respiratory
Turkey
Cow
Dog
Cat
Cat
Enteric
Enteric
Enteric
Enteric
Peritonitis
Pig
Human
Human
Human
Mouse
Parrot
Rabbit
Encepha IomyeIitis
Respiratory
Enteric
Demyelinating
Hepatitis
Enteric
Pleuritis
Pig
Rabbit
Enteric
Enteric
Rat
Rat
Pig
Respiratory
Adenitis
Enteric
Antigenic Interrelationships
Although data on the antigenic interrelationships among
the Coronaviridae are not yet complete, the viruses have
been tentatively placed into groups on the basis of cross­
reactivity in serological tests (Table 2).
The mammalian
■coronaviruses which have been classified so far fall into
two groups.
The avian coronaviruses, infectious brochitis
5
Table 2. Antigenic cross-reactions among coronaviruses
MAMMALIAN
Unclassified
Group I
Group 2
229EV
TGEV
CCV
FIPV
OC43V
MHV
RCV
SDAV
BCV
HEV
SD , SK
CV-777
FECV
HECV
RbCV
RbECV
TCV
PCV
AVIAN
IBV
virus (IBV) and turkey coronavirus (TCV), are apparently
unrelated to each other and to the other coronaviruses
(7,40,76).
The human coronaviruses
(HCV) have been placed
into separate antigenic groups based on the results of
enzyme-linked immunosorbent assay (ELISA)
(37),
polyacrylamide gel electrophoresis (PAGE) (55), and
Immunoelectrophoresis
(IEP) (54).
Group I contains HCV-
229E (229 E V ) and strains which can be propagated in cell
culture.
The organ culture virus OC43V and other strains .
that could not be cultivated in vitro are found in Group
2.
Viruses of the OC43 group cross-react with murine
hepatitis viruses A S 9 and JHM (A59V,
J H M V ) in ELISA (37).
These results are consistent with earlier studies
6
in which virus neutralization (VN) (40), complement
fixation
(CF) and gel diffusion (7) tests suggest a
relationship between the MHV's and several human
coronaviruses.
They also report a complete lack of cross­
reactivity between IBV and any of the other known
coronaviruses.
The rabbit coronavirus RbCV has been
reported to show cross-reaction with both 229EV and
OC43V (59).
A coronavirus-like agent isolated from
swine and designated CV-777 (44) was compared by immunoelectron microscopy and immunofluorescence to a number of
classified coronaviruses (12,45).
No cross reaction was
seen between CV-777 and IB V , T G E V , CCV, H E V , B CV or FIPV.
The morphological appearance is so far the only criterion
for inclusion of this isolate in the coronavirus family.
The hemagglutinating encephalomyelitis virus of swine
(HEV), previously placed in the coronavirus family on the
basis of its characteristic morphology, was shown to be
related to the OC43V group of human corona viruses in
hemagglutination inhibition (HAI), CF and VN tests
(30).
A two-way cross-reaction between HEV and OC43V was
demonstrated by all three methods.
Another interesting
addition to this antigenic group was recently reported by
Gerdes et al.
(5,20).
Coronaviruses SD and SK isolated
from the brains of two multiple sclerosis patients are
serologically related to A59V, JHMV, and OC43V.
7
The bovine coronavirus (BCV) cross-reacts in V N and
HAI tests with OC43V (21) and with HEV
(53).
Also,
neutralizing antibodies against BCV were detected in sera
from humans (61) and a number of animal
species
(52),
suggesting the possible existence of related viruses in
these
species.
The final members of this antigenic group (Group 2) at
present are the rat coronavirus
(RCV) and
sialodacryoadenitis virus (SDAV) which are closely related
to the murine hepatitis viruses
(76).
Group I of the mammalian coronaviruses contains the
transmissible gastroenteritis virus of swine (TGEV),
canine coronavirus (CCV), feline infectious peritonitis
virus (FIPV) and the human coronavirus 229E and related
strains.
229EV cross-reacts with the other three viruses
in immunofluorescence assay (IFA), although the reactions
are weak (43).
The existence of a canine coronavirus
(CCV) was first suspected when TGEV antibodies were found
in dogs that had never had contact with pigs (18).
Later
the CCV was i s o l a t e d from dogs (I) and the i s o l a t e indeed
proved to be closely related to TGEV in various
serological tests.
The two viruses can be differentiated
by reciprocal VN experiments in which homologous titers
are significantly higher than heterologous antibody levels
(49,78).
The FIPV also has a high degree of cross­
reactivity with TGEV and CCV in VN
(47,48,49,77,78) and
8
IFA (43,48,77) tests.
In cross-protection studies
conducted by Woods and Pedersen (77) pigs immunized with
FIPV did not develop detectable T G E V -neutraIizing
antibodies, but seemed to be protected to a certain extent
against challenge with TGEV.
On the other hand, dats
vaccinated with TGEV produced cross-reacting antibodies to
TGEV and F I P V , but were not protected against FIPV
challenge.
Cats given TGEV orally showed no signs of
clinical disease, but shed infectious v i r u s •and developed
VN antibodies
(48,77).
Molecular Immunology
There are few data available on the cross-reactivity
among the coronaviruses at the molecular level.
Several
investigators have reported on the antigenicity of
subviral components or individual polypeptides.
contain three major antigens:
Virions
the p e p lomeric glycoprotein
P, the nucleocapsid protein N and the envelope or membrane
glycoprotein(s ) M
(19,22,24,29,55,79).
The peplomeric
protein is apparently responsible for generation of
neutralizing antibodies.
Animals immunized with TGEV
surface projections or intact virions develop VN
antibodies, while subviral particles do not elicit
neutralizing antibodies
(19,24,49,55),
suggesting that
neutralizing antibodies raised during TGEV infection are
9
directed against virus peplomers.
Horzinek et al.
(29)
investigated antigenic relationships among homologous
structural proteins of T G E V , CCV and FIPV using ELISA, VN,
immuneprecipitation and immunoblotting techniques.. These
studies suggest reciprocal cross-reactivity of all three
major viral antigens.
The strongest antibody response was
directed at the envelope proteins, in contrast to the
previous work in which the response seemed to be toward
the peplomeric protein.
The polypeptides of 229EV and OC43 were compared by
PAGE (55) and quantitative IEP (54).
Patterns were
s i m i l a r for the two viruses but there was no c r o s s ­
reaction between them.
Once again the neutralizing
antibodies were directed against the pep Iomeric
glycoprotein (37,54,55).
Antigenic interrelationships among murine coronaviruses were investigated by Fleming et al. (16) using a
panel of monoclonal antibodies to JHMV.
The patterns of
cross-reactivity were different for each strain,
confirming that they are closely related but in fact
separate strains.
Bond et al.
(3) and Cheley et al.
(8)
also demonstrated a high degree of relatedness among MHV
strains by tryptic peptide mapping of proteins and RNA
blotting.
The polypeptides of corona viruses SD and SK were
compared by radioimmunoprecipitation to proteins of
10
OC43V and A59V (20).
Reciprocal cross-react ions, were seen
among, the major antigens of all
four ■viruses. • Brian et
a I . (26) used immunoblotting studies to determine the
cross-reactivity among homologous structural polypeptides
of BCV, OC43V and A59V.
Monospecific antiserum prepared
against, each of the three major antigens of BCV reacts
with the corresponding proteins of OC43V and A59V.
An
additional antigen found in BCV and OC43 virions is not
detectable in A59V.
This protein is apparently the
hemagglutinin and has no homolog in A59V, a nonhemagglutinating virus.
Vautherot et a I. (74,7 5)
investigated the antigenicity of BCV using a panel of
monoclonal antibodies.
They reported finding two
peplomeric antigens, with the smaller one associated with
hemagglutinating activity of the virus (75).
These
studies also detected common antigenic determinants among ■
BCV, M H V , OC43V and H E V ,
I chose four mammalian coronaviruses for this study,
two from each antigenic group.
They are murine hepatitis
virus A59, bovine coronavirus, canine coronavirus and
transmissible gastroenteritis virus.
Murine Hepatitis Virus AS 9
The murine hepatitis viruses are a group of closely
related virus strains with variable pathogenicity for
mice (8,9,16).
A59V causes an acute fatal hepatitis in
11
mice.
It is readily propagated in a number of tissue
culture cell lines and has been extensively studied.
Together with the other M H V ^s it is a p r o b l e m in mouse
breeding colonies.
Part of the importance of these
viruses is in their use as m o d e l s for studying the
pathogenesis of hepatitis, encephalitis and more recently,
demyelinating diseases such as multiple sclerosis
(20,76).
A5 9V has been we I !-characterized biochemically.
Virions contain three major structural proteins:
a
peplomeric glycoprotein of approximately 180 kd, a 50-60
kd nucleocapsid phosphoprotein and a heterogeneous
membrane glycoprotein of 23-26 kd (3,4,38,65).
When virus
particles are treated with trypsin, the P protein is
cleaved by trypsin to form two 90 kd polypeptides
(28,38,64,66).
The tryptic peptide maps for the 90 kd and
18 0 kd proteins appear to be identical (66).
The
intracellular proteins of A59V and the closely related
JHMV have been characterized in a number of laboratories
(3,4,51,58).
In addition to the structural
in the virion,
proteins found
there are several virus-specific
polypeptides found in infected cells with apparent
molecular weights of 57k, 54k, 3 9k and 37k.
Compared to
other corona viruses, the MHV's are relatively easy to
cultivate and assay and have been adapted to a number of
continuous cell lines.
12
Bovine Coronavirus
Bovine coronavirus was first characterized about ten
years ago by Mebus and co-workers
(42,60) as a
coronavirus-Iike e t i o logic agent of neonatal calf
diarrhea.
Symptoms begin 24-30 hours after inoculation,
and last 4-5 days.
calves (76).
B C V infection can be fatal in newborn
Like the parvo- and rotaviruses, BCV is a
major cause of neonatal calf diarrhea and is of economic
importance in the United States.
One of the problems with.
BCV studies is the difficulty of propagating and assaying
the virus in v i t r o . Most of the e a r l i e r work was done in
primary or other non-continuous cell lines.
Laporte et
a l . (35) reported in 1980 the use of a human
adenocarcinoma cell
titers of BCV.
line, HRT-18, for cultivation of high"
Vautherot (73) later developed a plaque
assay using this same cell line.
These developments have
greatly facilitated the biochemical and antigenic
characterization of BCV.
There is so far on l y one serotype of BCV, altho u g h Dea
et al.
(11) recently reported differences in counter-
immunoelectrophoresis and immunodiffusion patterns among
five BCV isolates.
These results are awaiting
confirmation, since the same five strains cross-reacted in
reciprocal VN experiments done earlier by the same group.
Only two precipitating antigens are detected by these
13
methods,
in contrast with the four antigens identified in
the LY-I38 strain of BCV by Hajer and Storz (23).
The
monoclonal antibodies prepared by Vautherot et a I. (74,75,)
detected only minor antigenic variation among BCV isolates
from various sources.
Two of the monoclonal antibodies
directed against a'p e p lomeric glycoprotein of the
immunizing (French) BCV strain failed to react with
isolates from the United States and Great Britain.
A
bovine respiratory isolate tested had the same reactions
.
as BCV with all of the monoclonal antibodies.
There is some disagreement as to the number and
character of BCV structural proteins.
Various workers
report from four to nine polypeptides in purified virions,
with molecular weights ranging from 23 to 190 kd
(23,26,33,36,62,74).
The most recent studies indicate
that the pepIomers are composed of a large (125-190 kd)
glycoprotein which is normally present as two smaller
subunits (33,74), and another glycoprotein of 105 kd
(74,75) or 140 kd (26) which is responsible, for the
hemagglutinating activity of the virus.
There is general
agreement that the nucleocapsid protein is a
phosphorylated' polypeptide of about 50 kd and a
heterogeneous glycoprotein of 23-25 kd is found in the
envelope matrix (23,33,36).
The intracellular non-
structural proteins of BCV have not been described.
14
Transmissible Gastroenteritis Virus
Transmissible gastroenteritis is an infectious
disease of pigs which is of major economic importance.
was first described by Doyle and Hutching in 1946
Tt
(14)
and was later shown to be caused by a member of the
^coronavirus family (41).
TGE is highly infectious and
has an 18-24 hour incubation period followed by diarrhea
and vomiting.
The symptoms are much less severe in adults
than in newborn animals, where the mortality rate can
approach 100%
(41,76).
serotype of TGEV.
There is probably only one
A number of strains have been isolated
in various parts of the world, but all are serologically
identical as tested so far (41).
C V -Ill,
a coronavirus­
like agent isolated recently from pigs with enteric
disease (12,44,45), does not cross react with TGEV
(or any
other established member of the Coronaviridae) and is
probably not a serotype of TGEV (45).
There has been much
research concerning the pathogenesis and prevention of
TGE.
A number of vaccines have been introduced,
are only partial Iy successful
but these
(17).
TGE virions contain three major structural proteins of
160-200,
50 and 28-30 kd (17).
The large glycoprotein is*
associated with the peplomers and elicits the production
of neutralizing antibodies (19).
TGEV-specified
intracellular proteins have not been described.
15
Canine Coronavirus
The canine coronavirus is closely related to TGEV;
indeed it is has been suggested that it, together with
F I P V , is a host-range mutant of TGE V rather than a
separate species (29).
et al.
CCV was first described by Binn
(I). This isolate,
designated 1-71, can be
experimental Iy transmitted to puppies, resulting in a
self-limiting course of gastroenteritis and dehydration
(31).
Adult dogs show no clinical signs of disease,
develop neutralizing antibodies to CCV.
but
From 62-87% of
kennel dogs are seropositive for CCV (76).
Canine coronaviruses is fastidious and has usually
been cultivated in primary or secondary cells
(1,18,29,31,
49). These cell lines vary in their susceptibility to CCV
strains and cytopathic effects (CPE) are not always seen.
Woods (78) reported the development of a feline cell
line
(FC) s u i t a b l e for growth of CC V as w e l l as for FIP V and
TGEV.
cells.
All three viruses cause CPE and form plaques in FC
A continuous canine cell line, A- 72, was
established by Binn et al.
(2) for propagation of a number
of viruses including CCV.
The biochemistry of CCV has been less studied than
that of other coronaviruses, perhaps because of the
difficulties in propagating the virus in vitro.
and Reynolds (18) have determined the polypeptide
Garwes
16
structure of purified virions.
structural proteins:
There are four major
gp204, pp50, gp32 and gp22.
The
first three correspond to the proteins of TGE V .
The 22 kd
glycoprotein apparently has no homolog in TGEV.
The
authors suggest that CCV may contain two membrane
glycoproteins.
However, the membrane glycoprotein is
knqwn to be heterogeneous in many of the studied
coronaviruses
(65,68),
so separate bands seen in gel
electrophoretograms may represent different stages of
gIycosy I ation rather than different polypeptide species.
There are as yet no reports of non-structuraI proteins
coded for by CCV.
Methods for Investigating Antigenic Relationships
One o b j e c t i v e of my research was to use these four
coronaviruses in a variety of immunological tests to
further the previous studies of their antigenic
relatedness and to extend the findings to the level of
individual structural and non-structural polypeptides.
addition to verifying antigenic relationships,
In­
these
studies would be helpful in increasing our understanding
of intracellular processing of virus—specific proteins.
The antigenic groups shown in Tab l e 2 are based
primarily on inter-species serological tests.
The
experiments used serum from one host animal to test virus
from another species.
The use of heterotypic serum' can
17
yield misleading results due to endogenous antibodies
directed against non-viral substances present in the
sample.
Also, there may be significant variations in
sample quality when using immune serum from infected
animals.
I wanted to conduct a study using a homogeneous
system with antibody from a single source.
Comprehensive
serological studies of this type require substantial
quantities of poly- and monospecific antibodies.
Large
volumes of hyperimmune ascitic fluid can be induced in
individual immunized mice by injection of Freund's
adjuvant (71) or sarcoma-180 tumor cells
(25,70).
In
these procedures mice are generally immunized with
purified virus particles to generate highly specific
antibody preparations.
However, in this case antibodies
are produced only against external structural components
of the immunizing virus.
A comprehensive investigation of
antigenic relationships at the molecular level should
include comparisons of virus-specified intracellular
polypeptides and internal virion proteins as well.
.I
wanted to develop a protocol for production of anti-viral
ascitic fluid in mice using infected-cell lysates as
immunogen.
Five criteria were established for development
of this system.
I.
The method should produce high-titer hyperimmune
fluids containing antibodies directed against
intracellular as well as virion proteins.
18
2.
The system should be homogeneous, with all
antibodies from a single
inbred mouse strain.
source,
e.g. the same
The methods should be
applicable to a wide variety of cel I-culture
adapted viruses.
3.
Large quantities of hyperimmune fluids should be
obtainable from individual mice.
4.
There should be a simplified protocol for
preparation of viral immunogen to obviate the
need for extensive purification.
5.
Mice immunized in this manner could also be used
for production of monoclonal antibodies using
hybridoma technology.
A panel of monoclonal and polyvalent antibodies
recognizing both structural and intracellular proteins
can be generated for use in molecular studies of antigenic
relationships and protein processing.
Experimental Design
I conducted this research in three stages.
The first
was the establishment and characterization of virus-cell
systems for each of the four coronaviruses chosen for the
study.
Continuous cell lines were tested for cytopathic
effect (CPE), virus yield, and ability to quantify virus
by plaque titration. The second phase of this project was
the preparation of polyvalent and monoclonal antibodies
19
against each virus.
Anti-viral hyperimmune ascitic fluids
were prepared in mice and monoclonal antibodies produced
by cell fusion techniques.
Finally, serological and
molecular immunology methods were used to assess antigenic
cross-reactions among homologous intracellular
polypeptides of the four viruses.
Anti-viral ascitic
fluids and monoclonal antibodies were used in reciprocal
cross-neutralization, immunofluorescence and
immunoprecipitation tests to define and compare the
antigens of the viruses.
20
MATERIALS AND METHODS
Chemicals and Media
Chemicals and reagents were obtained from Sigma
Chemical Co. unless otherwise stated in the text.
Radioisotopes were obtained from New England Nuclear
Corporation.
Plastic dishes for cell culture were
obtained from Nunc, Denmark.
Media used for cell culture
were purchased in powder form from Irvine Scientific
(Santa Ana,
CA).
(Logan, U T ).
Sera were obtained from Sterile Systems
DME-O consisted of Dulbecco's Modified
Eagle's medium supplemented with 200 Units/ml of
penicillin G and 25 ug/ml of streptomycin.
DME-IO was
prepared by adding I volume of calf serum to 10 volumes of
DME-0.
DME-20 was prepared by adding 2 volumes of fetal
bovine serum and I ug/ml of amphotericin B to 10 volumes
of DME-0.
DME-2 contained 2% fetal bovine serum and was
supplemented with 10 m M 3-(N-morpholino)propanesulfonic
acid (MOPS),
10 mM N-tris-(hydroxymethyl)methy1-2-amino-
ethanesulfonic acid (TES), and 10 m M N-2-hydroxyethylpiperazine-N'-2-ethanesul fonic acid
(HEPES ).
HAT medium
for hybridoma cultures was prepared by adding 1% of 100X
21
stock solutions of HT (10 mM hypoxanthine,
thymidine)
1.6 mM
and A (0.04 mM aminopterin) to DME-20.
Virus Strains and C e l l Lines
The murine coronavirus A59V was obtained from Dr.
Lawrence Sturman.
Bovine coronavirus (BCV) was obtained
from the American Type Culture Collection (ATCC VR-874),
as were the canine coronavirus (ATCC VR-809) and the
transmissible gastroenteritis
virus (ATCC VR-743).
A59V was propagated in murine 17CL-1 cells,
a
spontaneously transformed continuous cell line cloned from.
BALB/c-derived 3T3 fibroblasts'and obtained from Dr.
Sturman (63).
A continuous human adenocarcinoma cell
line
obtained from Dr. David Brian, HRT-I8 (35,69), was used
for cultivation of BCV.
TGEV was propagated in swine
testicle (ST) cells, a continuous cell line established by
McClurkin and Norman (39) and obtained from Dr. Brian.
Canine coronavirus was adapted to the continuous A - 72 cell
line (ATCC CRL-1542) established by Binn. et a I (2).
murine sarcoma cell
The
line S180 (ATCC TIB-66) was used for
induction of ascites in immunized mice.
The BALB/c
myeloma line P3/NSl/l-Ag4-l (NS-I) is a non-secreting
clone of P3X63Ag8.
Institute.
NS-I cells were obtained from the Salk
Al I cell lines were maintained in DME-IO
except for NS-I cells, which were passaged in DME-20.
22
Preparation of C e l l Lysates
Virus-infected cell lysates were used as stock virus
and to immunize mice for antibody production.
Cell
monolayers in 10 cm plastic tissue culture dishes were
infected with virus at a multiplicity of infection (MOI)
of 1-5.
After an adsorption period of I hr at room
temperature the inoculum was removed and 5 ml of DME-2 was
added.
The cultures were incubated at 37°C and harvested
when extensive cytopathic effect was seen.
TGEV and AS 9V
were harvested at 18-24 hr post-infection .(HPI);
CCV at approximately 48 HPI.
BCV and
Mock-infected cell cultures
were prepared in the same manner, using an inoculum of
DME-2 instead of virus.
Cells were disrupted by one cycle
of freeze-thawing at -70°C.
The resulting lysate was
scraped with a rubber p o l i c e m a n and sonicated 1-2 min in a
Heat Systems Sonicator (model W-225R) using a cup probe at
70% power.
The lysates were clarified by centrifugation
at 1500 x g for 5 min and stored at - 7 0°C.
Plaque Assay
Virus stocks were cloned twice by plaque selection and
titered by plaque assay on the appropriate cell monolayer.
C e l l s (I x IO^ in D M E - I O ) were seeded into p l a s t i c sixwell dishes and incubated at 37°
. ,
C overnight.
Monolayers
were washed with pre-warmed DME-O and infected with 0.2 ml
23
of 10-fold dilutions of virus in DME-2.
After an
adsorption period of I hr at room temperature the inoculum
was r e m o v e d and 2 ml of an o v e r l a y consisting of 0.75%
agarose (Sigma Type I ) in DME-2 was added.
Plates were
incubated at 37°C for 2 days (TGEV, A59V) or 3-5 days
(BCVfC C V ).
plaques,
If staining was required for visualization of
the cells were fixed with 2% glutaraIdehyde.
After fixation for I hr or more, the agarose o v e r l a y was
removed.
The fixed cells were stained with 1% aqueous
crystal violet for 30 min, washed several times
water and air-dried at room temperature.
with
Plaques were
counted and the virus titers expressed as plaque-forming
units per ml
(PFU/ml).
Immunization of Mice
The immunogen preparation consisted of mock or virusinfected cell lysates emulsified with an equal volume of
complete Freund's adjuvant (CFA).
Adult (8-12 w k ) Balb/c
mice were given three intraperitoneaI (ip) injections of
immunogen, 0.3 ml each, at weekly intervals.
A final
booster injection consisting of 0.3 ml cell lysate without
adjuvant was given one week later.
Each inoculation
contained IO^ to 10® plaque-forming-units (PFU) of virus.
24
Harvesting of Serum and Ascitic Fluid
Ascites were induced in immunized mice by injection of
Sarcoma-180 cells
(25,70).
Normal mouse ascitic fluid was
prepared by injection of unimmunized mice with S-180
cells.
One day after the final booster immunization, mice
were injected ip with 0.3 ml of sterile phosphate-buffered
saline
(PBS: 140 mM N a C I, 2.7 mM K C I, 9.4 mM N a 3 HPO4 , 1.5
mM KH2PO4, 0.9 mM CaC l 2 , 0.5 mM M g C l 2 -GH 3 O, I mM NaN3 )
containing 5 X IO6 S-180 cells.
The abdomens became
markedly distended within 10-15 days, at which time the
accumulated fluid was removed by paracentesis using an 18gauge needle.
Surviving mice continued to accumulate
ascitic fluid and were tapped again every 2-3 days. It was
p o s s i b l e to obtain as much as 50 ml of fluid from an .
individual mouse treated in this manner.
Sera were
obtained from the same mice by tail bleeding.
Three mice
were immunized with each virus preparation and the fluids
pooled.
Ascitic fluids were refrigerated overnight and
c entrifuged at 800 x g for 5 min to remove c e l l s and
debris.
Absorption of Serum and Ascitic Fluid
Prior to use in serological studies the sera and
ascites were absorbed with uninfected cells to -remove
antibodies directed against non-viral components.
25
Confluent monolayers of the cell
lines used to propagate
each virus were grown in 10 cm plastic dishes.
The
monolayers were washed carefully once with PBS and once
with absolute methanol, fixed 2 min in methanol, dried
with compressed air and stored in plastic bags at -70°C
until needed. Immediately prior to use the plates were
rehydrated by washing once with PBS. For absorption,
I ml
of serum or ascitic f l u i d was layered on a 10 cm dish
containing the appropriate fixed cells and incubated
overnight at 4°C.
Fluids were absorbed twice in this
.manner and stored frozen (-70°C) until used.
Preparation of Monoclonal Antibodies
Mice were immunized as described above.
Spleens were
removed from immunized mice 3-4 days after the final
booster injection. Immune spleen cells (3 x IO^ cells)
were fused with 5 x IO7 NS-I cells using 50% polyethylene
glycol
(mol. w t . 1000, Sigma cat. no. P3515).
Fused cells
were diluted in HAT medium and seeded into 96-well plates
containing 10^ mouse peritoneal macrophages per well
(13).
Incubation and maintenance of the hybrids was carried out
according to the microculture protocol of de St. Groth and
Scheidegger (13).
Culture fluids from growing colonies
were screened for anti-viral antibodies by immuno­
fluorescence.
Cells from positive wells were cloned twice
by limiting dilution in 96-well plates.
Supernatant
26
fluids from the cloned hybridoma cell lines were used in
neutralization, immunofluorescence and immunoprecipitation
tests.
Alternatively, ascitic fluids containing high
concentrations of anti-viral monoclonal antibody were
prepared by injecting 1.5 x IO^ hybridoma cells into
pristane
(2,6,10,14-tetramethy!pentadecane)
treated mice.
The mice were injected ip with 0.5 ml of pristane at least
one week prior to the injection of hybridoma cells.
Immunofluorescence Assay
A microculture immunofluorescence assay (IFA)
developed by Robb (50) was used to assess antibody
activity of ascitic- fluids and sera and to screen
hybridoma supernatant fluids for specificity.
Cells were •
suspended in DME - 2 at a concen t r a t i o n of 6 x IO^ c e l l s per
ml and mix e d with the corres p o n d i n g virus in a ratio of 9
parts cells to I part virus or DME-2 (mock-infected).
The
infected and mock-infected cells were seeded (10 u 1/we 1 1 )
into 60-well Terasaki plates using a Hamilton repeating
dispenser such that each plate contained 30 wells of mock
and 30 wells of virus-infected cells.
The plates were
incubated at 37°C for 8 hr (TGEV, A59V) or 18-24 hr (BCV,
C C V ), washed and fixed with methanol as described above
and stored at -70°C.
For indirect immunofluorescence
staining the plates were thawed and rinsed once with PBS.
Ten microliters of hybridoma culture medium, diluted serum
27
or ascitic fluid was added to each well.
After 30 min at ■
room temperature the plates were washed 4 times with PBS
and 10 uI of fluorescent isothiocyanate (FITC)-conjugated
goat-anti-mouse immunoglobulin (Antibodies Inc. cat. no.
2146) was added to all wells.
The plates were incubated
an additional 30 min, washed 4-5 times with PBS and
observed
using an Olympus IMT inverted microscope with
reflected fluorescence accessories.
Plaque Reduction Test
A plaque reduction test was used to determine the
virus neutralizing (VN) activity of the hyperimmune sera
and ascitic fluid.
Serial dilutions of antibody were made
in D M E - 2 , mixed with an equal volume of virus suspension
(approximately 600 PFU/ml) and incubated 30 min at 37°C.
Confluent monolayers of the corresponding cell
line in 6-
w e 11 dishes were inoculated with 0.5 ml of the virusantibody suspension and the plaque assay completed as
described above.
The number of plaques in each antibody-
containing well was compared to the virus control wells to.
determine the percent p.laque reduction (PR).
Computer
programs were used to p l o t PR vs the reciprocal of the
antibody dilution (BPS Business Graphics, Cambridge,
MA)
and to calculate the dilution giving 50% plaque reduction
(Omicron Plotrax,
Engineering Sciences, Atlanta, GA).
The
VN titer of the antibody preparation was reported as the
28
dilution resulting in 50% plaque reduction.
The percent
PR of each virus by a 1:10 d i l u t i o n of normal mouse
ascitic fluid (NMA) was determined as a negative control.
An antibody preparation was considered to have no
detectable neutralizing acitivity against a virus if a
1:10 d i l u t i o n r e s u l t e d in a lower PR than the NMA.
Monoclonal antibody supernatants were tested for
neutralizing activity by a modification of the plaque
reduction assay.
Virus suspensions were diluted to
approximately 300 PFU/ml and mixed with an equal volume of
undiluted hybridoma supernatant medium.
After incubation ■
for 30 min at 37°C, 0.5 ml of the virus/antibody mixture
was inoculated onto washed cell monolayers and the plaque
assay completed as described.
was used as a negative control.
HAT medium without antibody
Monoclonal antibodies
were considered to be neutralizing if the plaque reduction
was greater than 90%.
Radio IabeIing of Intracellular Proteins
Confluent cell monolayers in 6-cm plastic dishes were
inoculated with stock virus at an MOI of I to 5 and
incubated at 37°C.
At various times post-infection the
medium was removed and replaced with 0.7 ml of methioninedeficient DME-2 containing 200 uCi/ml of ^ S -methionine.
Cells infected by A59V or T G E V 'were labeled at 8 hr post­
infection (HPI), and BCV and CCV were labeled at 12 HPI.
29
After a I hr l a b e l i n g period the med i u m was r e m o v e d and
the cells were washed twice with DME-0.
Cells were lysed
with 0.3 ml of buffer BlO (IOmM Tris-HCl pH7.4,
MgC 1 2 , 0.5% NP-40,
5 mM
0.1% SDS, 1% Aprotinin, 50 ug/ml
ribonuclease A, 50 ug/ml deoxyribonuclease) for 5 min on
ice.
The lysates were harvested and stored at -2 0°C.
Immunoprecipitation of Virus-specific Proteins
Intracellular proteins were immunoprecipitated with
hyperimmune ascitic fluid as described previously (4),
with minor modifications.
Radiolabeled virus and mock-
infected cytoplasmic lysates were prepared as described
above.
Cell
lysates
(15 ul samples) were incubated with 5
uI of h yperimmune ascitic fluid or serum, or 20 ul of
hybridoma supernatant medium in 0.5 ml of Radioimmunpre­
cipitation (RIP) buffer (50 m M Tris-HCl pH 7.4,
N a C l , 5 mM EDTA, 0.2% NP-40,
15 0 mM
0.05% SDS, 1% Aprotinin,
0.02% sodium azide) for I hr at 0°C.
Immune complexes
were precipitated with 50 ul of 10% StaphIylococcus aureus
(Cowan) prepared by the method of Kessler (32) for I hr at
0°C.
The bacteria were pelleted by centrifugation in an
Eppendorf microcentrifuge at 6500 x g for 15 sec and
washed 3 times in RIP buffer.
The proteins were eluted
with 25 ul of 20 m M di t h i o t h r e i t o I ,' I % SDS for 15 min at
room temperature and 5 min at 60°C.
After centrifugation
at 6500 x g for 3 min the supernatants were removed, mixed
30
with an equal volume of SDS-PAGE diluent (120 mM Tris-PO^
pH 6.7, 1% SDS, 40% glycerol,
at -20°C.
0.02% phenol red) and stored
Controls consisting of lysates precipitated
with normal mouse ascitic fluid or S. aureus without
antibody were prepared in the same manner.
SDS-Polyacrylamide Gel Electrophoresis
Proteins were e l ectrophoresed on 8% polyacrylamide
slab gels as described by Laemmli and Favre (34), except
that the resolving gel was supplemented with 0.5% (wt/vol)
linear polyacrylamide (BDH Chemical Ltd.).
Following
electrophoresis the gels were fixed overnight in 5%
trichloroacetic acid (TCA).
Proteins were detected by
staining with brilliant blue G (15) or by impregnating the
gels with 10% (wt/vol) 2,5-diphenyloxazole (PPG) in
dimethyl sulfoxide (DMSO) followed by drying and exposure
to preflashed Kodak XAR-2 x-ray film at -70°C (6).
The molecular weights of virus-specific proteins were
determined from their distance of migration in slab gels
relative to those of standard proteins of known molecular
weight
(56).
Proteins used as markers were thyroglobuI in
(200 kd), beta-galactosidase (115 kd), phosphoryIase B
(97.4 kd), bovine serum albumin (66 kd), ovalbumin (45 kd)
and carbonic anhydrase
(29 kd).
31
RESULTS
Establishment of Virus-Cell Systems
The four coronaviruses chosen for this study were
adapted to cell culture as described in Materials and
Methods.
Cell lines were tested and selected on the basis
of cytopathic effect (CPE), virus yield and plaque assay
characteristics.
The murine hepatitis virus A S 9 was propagated in
17CL-1 cells, a spontaneously transformed derivative of
the BALB/3T3 cell line.
This virus-cell system had been
previously established in our laboratory (4), and high
titer (10® PFU/ml) virus-infeeted cell lysates were
available.
A59V caused extensive cytopathic effect in
I7CL-I cells, including syncytia formation and detachment
of cells from the monolayer.
Syncytia formation began at
5-6 hr post-infection (HPI), and cell destruction was
usually complete by 24 HPI.
Virus titers were determined
by plaque assay, with plaques clearly visible at 48 HPI.
The bovine enteric coronavirus (BCV) was adapted for
growth in a continuous human adenocarcinoma cell
HRT-18 (35,69).
line,
Several blind passages were done before
any evidence of CPE was seen.
After adaptation to HRT-18 '
cells, CPE was observed beginning at about 24 HPI and
32
consisted of rounding up and vacuolization of the cells.
The CPE- began in discrete foci but spread to the entire
monolayer by 48 HPI, although there was very little cell
detachment.
Infected-cell
lysates for stock virus were
harvested at 48 HPI or later, when CPE was complete.
A.
plaque assay developed by Vautherot (73) was used to assay
BCV in HRT-I 8 cells, and virus stocks with titers of at
least IO7 PFU/ml were prepared.
The virulent Miller strain of transmissible
gastroenteritis virus (TGEV) was propagated in swine
testicle (ST) cells (39).
After initial isolation and
plaque purification in ST cells, only low titer (IO5
PFU/ml) virus was produced, and CPE took 4-5 days to
develop.
Further adaptation was carried out by Dr.
Andreas Luder.
Repeated passages of virus at a high
multiplicity of infection (MOI) resulted in cell lysates,
with titers of greater than IO8 PFU/ m l . . Cells infected
with this stock virus began to round up. and detach within
12 hr, and cell lysis was n e a r l y compl e t e by 24 RPI.
A
plaque assay was established for TGEV in ST cells.
Plaq u e s 2-3 mm in diameter were v i s i b l e within 4 8 hr.
Canine coronavirus
A - 72 cell
(CCV) was adapted to the continuous
line established by Binn et al. (2).
Cytopathic
effects including syncytia formation and lysis of infected
cells were seen beginning with the third, passage of CCV in
these cells.
Syncytia formation began at about 12-15 HP I ,
33
and cell lysis was usually complete in 48 hr.
assay was developed with some difficulty,
A plaque
since the A-72
cells were fragile and tended to lyse spontaneously after
several days incubation.
Plaques appeared in 4-7 days and
usually required crystal violet staining for optimum
visualization.
I was unable to produce CCV stocks with
titers greater than 3 x IO6 PFU/ml despite repeated
adaptation passages.
However,
these titers were adequate
for my studies.
Once satisfactory stocks of the four coronaviruses
were prepared, a time course experiment was conducted to
determine the optimum time period for detection of intra­
cellular viral protein synthesis (Figure I).
The peak of
protein synthesis in A S 9V infected cells was from 6-9 HPI,
approximately the same as for TGEV-infected cells.
CCV
protein synthesis reached a maximum at about 15 HPI.
Intracellular BCV protein synthesis continued over a
longer period (from about 12-21 H P I ), probably because
there was very little cell destruction before 24 HPI.
Preparation and Testing of Anti-Viral Ascitic Fluid's
Hyperimmune ascitic fluids' were induced in mice
according to the protocol described in Materials and
Methods.
Mice were immunized with virus-infected cell
lysates to produce antibodies against structural and
intracellular proteins.
Induction of ascites with
34
7000-t
6500600055005000—
4500-
Q
4000-
^
3500-
^
3000-
D
Ol
q
2500-
200015 0 0 -
1000500-1
°r
0
5
10
15
20
25
30
35
40
HOURS POST-INFECTION
Figure I. Time course for coronavirus protein synthesis
in cells infected with A59V (+ ), BCV (□), CCV (o) and
TGEV(B). Virus and mock-infected cells were pulse-labeled
with -^S-methionine at various times post-infection as
described in Materials and Methods. Cell lysates were
immunoprecipitated with homologous anti-viral ascitic
fluid, and the amount of virus-specific protein determined
by liquid scintillation counting. Data are expressed as
counts per minute (CPM) per 10 u I sample.
35
Sarcoma-180 cells resulted in substantial fluid
accumulation in individual mice.
Up to 40 ml of fluid
could be obtained from a single mouse.
Antibody-
containing fluids from three mice were pooled for each
virus to compensate for individual variations in immune
response.
Fluids were absorbed with uninfected cells to
remove non-viraI antibodies and tested for specificity
against the homologous virus by IFA.
A comparison of
absorbed and unabsorbed anti-A59V and anti-BCV ascitic
fluids is shown in Figure 2.
A59V was propagated in a
syngeneic cell line, so there should be no antibodies
against cellular antigens in the hyperimmune fluid.
The
background (uninfected cell) fluorescence was about the
same for absorbed or unabsorbed anti-AS9V (Figure
2A,2B,2C,2D).
In the BCV IFA, it was impossible to
distinguish virus-infected from uninfected cells if
unabsorbed fluid was used (Figure 2E,2G).
With absorbed
ascitic fluid the background fluorescence was greatly
reduced and the virus-infected cells were clearly visible
(Figure 2F,2H).
The results for the other two viruses
were e s s e n t i a l l y the same as in the BCV assay.
The anti-
CCV ascitic fluid had high background fluorescence and
required absorption to visualize infected cells.
anti-TGEV fluid had a lower background, although
absorption improved contrast between infected and
uninfected cells
(data not shown).
The
36
A N T I - V I R A L ASC I T I C FLUID
A59 V-infected
1 7 C L - 1 cells
mock-infected
1 7 C L - 1 cells
BCV-infected
H R T - 18 cells
mock-infected
H R T - 18 cells
Figure 2. Comparison of unabsorbed and absorbed
hyperimmune ascitic fluids. HRT-I8 or 17CL-1 cells were
infected with BCV, A59V, or mock-infected. Immuno­
fluorescence assays using homologous absorbed and
unabsorbed anti-viral ascitic fluids were conducted as
described in the text. A phase contrast micrograph of the
same field of cells is shown beside each
immunofluorescence micrograph.
37
To further test the effectiveness of the hyperimmune
ascitic fluids, virus neutralizing titers were determined
■
and compared to those of serum from immunized mice.
Virus
neutralization (VN) curves using serum and ascitic fluid
against each of the four viruses are shown in Figures 3
and 4.
The VN titers of the corresponding serum and
ascitic fluid were similar (less than a two-fold dilution
apart) for A59V, BCV and CCV.
The anti-TGEV serum was
obtained from a different mouse than the ascitic fluid and
had a much lower anti-viral activity.
The ascitic fluids
also compared favorably to serum in immunofluorescence
assays
(not
shown).
Intracellular Virus-specific Proteins
Radio labeled intracellular proteins from virusinfected cell lysates were immunoprecipitated using
hyperimmune ascitic fluid.
Figure 5.
The results are shown in
Seven to nine intracellular viral proteins were
precipitated from A S 9V-infected I7CL-1 cells (lane I).
The apparent molecular weights of these polypeptides were
155, 113, 60, 56, 50, 45, and 23-26 kd.
The 23-26 kd
bands were heterogeneous.
A smaller 18 kd protein was
seen in some preparations.
BCV-infected'HRT-18 cells
contained seven viral proteins with molecular weights of
141-152k, 106-i15k,
100k,
60k,
48k,
33k and 26k (lane 3).
The anti-CCV ascitic fluid precipitated four viral
38
20000
o
1B00016000-
^
14000-
5
12000 -
10000-K
O
6 00 0-
O
4 000-
2000-
MHV-A59
Percent Plaque Reduction
5000-#•
5
4500-
4000-
> * 3500-
-O 3 000-
<
2500-
1500-
O
iooo-
Percent Plaque Reduction
Figure 3. Comparison of virus-neutralizing activity of
anti-AS9V and BCV serum and ascitic fluid. Serial
dilutions of homologous anti-viral serum (+) or ascitic
fluid (x) were tested for neutralizing activity against
A59V and BCV in plaque reduction tests as described in
Materials and Methods.
The reciprocal of the antibody
dilution was plotted against percent plaque reduction.
39
< soo-
Percent Plaque Reduction
1000C-¥
~
0000-
D
7 00 0-
to
3 000-
®
1000-
TGEV
Percent Plaque Reduction
Figure 4. Comparison of virus-neutralizing activity of
anti-CCV and TGEV serum and ascitic fluid.
Serial
dilutions of homologous anti-viral serum (+) or ascitic
fluid (x) were tested for neutralizing activity against
CCV and TGEV in plaque reduction tests as described in
Materials and Methods.
The reciprocal of the antibody
dilution was plotted against percent plaque reduction.
40
I
2
3
7
4
8
186 "» |
135
115^P
92
81
11*
Figure 5. Intracellular virus-specific proteins of
coronaviruses A59V, B C V , CCV and TGEV. Virus and mock
infected cells were labeled with ^S-methionine as
described in the text.
Cell lysates were immunoprecipitated with homologous anti-viral ascitic fluid and
analyzed by SDS-PAGE on 8% slab gels.
Viral proteins are
indicated with their approximate sizes in kd.
The figure
shows (I) A59V or (2) mock-infected 17CL-1 cells, (3) BCV
or (4) mock-infected HRT-I8 cells, (5) CCV or (6) mockinfected A-72 cells, and (7) TGEV or (8) mock-infected ST
cells. The lower portion of Lanes 5 and 6 was overexposed
to visualize the smaller proteins.
41
polypeptides of 185, 92, 47-50 and 30 kd (lane 5).
kd protein was sometimes detected.
A 40
The intracellular,
proteins of TGEV were 186, 135, 92, 81, 77, 48-50, 41 and
29-30 kd (lane 7).
T h e s e -intracellular proteins were reproducibly
detected in several experiments using different cell
lysates and in some cases different conditions of
immunoprecipitation including heating and alkylation of
samples.
The polypeptide profiles were essentially the
same in each experiment.
Serological Interrelationships
■'The four mammalian corona viruses used in these studies
are from two different antigenic groups.
CCV and TGEV
b e l o n g to Group I, and A S 9V and BC V are members of Group 2
(Table 2; 68,76).
These relationships were confirmed by
reciprocal cross-immunofluorescence, neutralization and
immunoprecipitation tests.
The results of an IFA using homologous and
heterologous ascitic fluids are shown in Figure 6.
It is
clear that A S 9V and BCV shared antigenic determinants,
did T G E V and CCV.
as
The results of an IFA are not
quantitative and do not give information about the degree
of cross-reaction between related viruses or the viral
components responsible for the relationship.
42
H Y P E R I MM U N E A S C I T I C FLUID
A 59V
BCV
CCV
T GEV
A59 V -infe c te d
1 7 C L - 1 cel l s
BC V - i n f e c t e d
H R T - 1 8 cells
CCV-infected
A - 7 2 cells
TGE V - i n f e c t e d
ST cells
Figure 6. Antigenic cross-reactions among coronaviruses
by immunofluorescence.
Cells infected with A59V, B C V , CCV
or TGEV were analyzed by immunofluorescence assay using
homologous and heterologous anti-viral ascitic fluids as
described in the text.
43
Table 3 contains the VN titers for homologous and
heterologous hyperimmune ascitic fluids against all four
viruses.
Homologous titers were higher in all cases; but
cross-neutralization was seen between A59V and BCV, and
between CCV and TGEV.
The data suggest that the antigenic
relationship between TGEV and CCV may be closer than that
of BC V and A59V.
In the latter case the c r o s s ­
neutralization titers were much lower than the VN titers
of the homologous antibody.
Table 3.
Virus
A5 9V
BCV
CCV
TGEV
VN a titers of hyperimmune ascitic fluids.
anti-A59V
3095
22
0
0
anti-BCV
14
1440
0
0
anti-CCV
anti-TGEV
Ob
0
71
575
0
0
53
3695
aReciprocal cross-neutralizations were conducted with
homologous and heterologous ascitic fluids as described in
Materials and Methods.
The numbers given are the
reciprocals of the dilutions resulting in 50% plaque
reduction.
^No detectable neutralizing activity.
Cross-immunoprecipitations of virus-infected cell
lysates with homologous and heterologous ascitic fluids
were done in order to examine the antigenic relationships
among the viruses at the molecular level.
The results of
SDS-PAGE of the immunoprecipitates are shown in Figures 710.
TGEV and CCV shared antigenic determinants on all
44
MOCK A 5 9 V
BCV
CC V TQEV CONT NMA
Figure 7.
Immunoprecipitation of intracellular MHV-A59
proteins by homologous and heterologous hyperimmune
ascitic fluids.
S-methionine-1abeled virus or mockinfected cell lysates were prepared as described in
Materials and Methods and imrnunopreci pita ted with antiA59V, B C V , CCV or TGEV ascitic fluid.
The immunoprecipitates were analyzed by SDS-PAGE.
Viral proteins
are indicated with their apparent molecular weights
(x IOjs). Lane I(MOCK) shows mock infected 17CL-1 cells
with anti-A59V. Lanes 2-5 show A59V-infected cells with
anti-A59V, anti-BCV, anti-CCV, or anti-TGEV.
Lane 6 (CONT)
shows A59V-infected cells, no antibody.
Lane 7 (NMA) shows
A59V-infected cells with normal mouse ascitic fluid.
45
MOCK
BCV
A59V
CCV
TGEV
M m
60
►
50 *»
Figure 8. Immunoprecipitation of intracellular BCV
proteins by homologous and heterologous ascitic fluids.
Immunoprecipitations were carried out as described in the
legend to Figure 6. Lane I shows mock-infected HRT-18
cells with anti-BCV.
Lanes 2-5 show BCV-infected HRT-18
cells with anti-BCV, anti-A59V, anti-CCV or anti-TGEV.
46
MOCK
CCV
TGEV
A59V
BCV
CONT
NMA
Figure 9. Immunoprecipitation of intracellular proteins
of CCV by homologous and heterologous ascitic fluids.
Immunoprecipitations were carried out as described in the
legend to Figure 6. Lane I shows mock-infected A-72 cells
with anti-CCV.
Lanes 2-5 show CCV-infected A-72 cells with
anti-CCV, anti-TGEV, anti-A59V or anti-BCV. Lanes 6 and 7
show CCV-infected A-7 2 cells with (6) no antibody, or (7)
normal mouse ascitic fluid.
47
Figure 10. Immun op reel pi tat ion of intracellular TGEV
proteins by homologous and heterologous ascitic fluids.
Immunoprecipitations were carried out as described in the
legend to Figure 6. Lane I shows mock-infected ST cells
with anti-TGEV.
Lanes 2-5 show TGEV-infected ST cells
with anti-TGEV, anti-CCV, anti-A59V or anti-BCV. Lanes 6
and 7 show TGEV-infected ST cells with (6) no antibody, or
(7) normal mouse ascitic fluid.
48
major viral proteins (Figures 9 and 10).
The 30 kd
protein of TGEV was precipitated by anti-CCV ascitic
fluid.
The band was visible on the original film although
it cannot be seen in the photo g r a p h in Figure 10 (all four
bands are shown in the cross-immunoprecipitation in Figure
12A, lane I).
The films were sometimes overexposed to
visualize the low molecular weight polypeptides, which
resulted in poor resolution of the large proteins (e.g.
the P proteins.
Figures 7,9,10).
The apparent recognition
of the 50 kd nuc I eocapsid protein of TGEV by all four
hyperimmune ascitic fluids (Figure 10) was probably due to
a non-immune binding of proteins to Sjl aureus (Cowan)
bacteria during immunoprecipitation.
Only the anti-TGEV
and anti-CCV fluids reproducibIy precipitated this
protein.
The non-specific binding of proteins to S.
aureus has been a recurrent problem in immunoprecipitation
studies (Control
lanes on Figures 6,8,9; Bond et. a I.,
unpublished data;
J. Leibowitz, personal communication).
Homologous antigens of BCV and A59V also showed cross­
reactions (Figures
7 and 8). The 115 kd protein of BCV was
not precipitated by anti-A59V, and some of the intra­
cellular polypeptides were poorly recognized by the
heterologous antibody.
The low molecular weight proteins
were usually precipitated by anti-BCV and anti-A59V
ascitic fluids, although they could not be seen on this
*
49
gel.
No reproducible cross-reactions were seen between
either AS9V or BCV and CCV or TGEV.
Antigenic Analysis using Monoclonal Antibodies
Monoclonal antibodies (MAbs) were prepared against
A59V, BCV, CCV and TGEV using h'ybridoma technology.
A
total of thirty-nine positive (by IFA) cultures was
detected among the four fusions:
eleven from A59V,
from C C V , three from BCV and seventeen from TGEV.
eight
More
than eighty positive clones were isolated from these
original cultures.
The clones were tested for virus
specificity and cross-reactivity by IFA, and eighteen
stable antibody-producing hybridoma lines were selected
for further examination.
The polypeptide specificity of
the MAbs was determined by immune-precipitation, and each
was tested for VN activity by the plaque reduction test.
A summary of the results is shown in Table 4.
Representative results of the immunoprecipitations
using monoclonal antibodies are shown in Figure 11.
Several of the MAbs precipitated more than one polypeptide
(see Table 4).
Each MAb was tested in two or three
independent immunoprecipitations.
Some of the minor
polypeptides evident in the original data are not visible
in the figure.
None of the MAbs showed any activity
against mock-infected cells by IFA or by
immunoprecipitation.
50
Table 4.
MAba
Characteristics of monoclonal antibodies.
Polypeptide specificity^
Structure I Intracellular
A24
AS
A4 3
Al?
A2 7
A74
B2 4
C49
C45
N
P
P
P
N
P
N
N
N
N
N
N
N
N
N
P
N
N
C39
C115
T20
T39
T40
T4 I
T22
T6
T7
(60)
(155)
(155)
(155)
(60)
(155)
(185)
(50)
(50)
VNd
50,57,160
132
100,132 ■
160
(60)
(50)
(50 )
(50 )
(50)
(50)
(50 )
(50)
(50)
Crossreactions^
40,92,190
40,42,52,190
42,52,92,190
40,77,81,92,190
40,81,92,190
40,77,81,92,190
30,150,190
90,135
40,47,77,81,92
40,47
- A59V
TGEV
TGEV
TGEV
TGEV
CCV
CCV
CCV
CCV
CCV
CCV
+
+
+
- ■
+
-
XVNe
NDf
ND
-
ND
ND
ND
ND
+
ND
ND
aMonoclonal antibodies were designated with a number
preceded by a letter identifying the immunizing virus.
A-A5 9V, B-BCV,- C-CCV, T-TGEV
^Polypeptides are listed by size in ki Ioda!tons.
cAs determined by immunofluorescence assay.
^Virus neutralization activity as determined by
plaque reduction.
eCross-neutralization by viruses other than the
immunizing virus as determined by plaque reduction.
^Not done.
Six MAbs with specificities for A59V polypeptides were
isolated.
Four of these were directed against the
peplomeric glycoprotein (Figure 11, lanes 2 and 5), and
51
Figure 11. Analysis of monoclonal antibody specificities
by immunoprecipitation. Radiolabeled virus-infected cell
lysates were immunoprecipitated with monoclonal antibodies
(MAbs) as described in Materials and Methods and the
polypeptides separated by SDS-PAGE in 8% slab gels.
Major
viral proteins are indicated with their apparent sizes in
kd.
Lanes 1-5 show A59V-infected cells immuneprecipitated
with (I) anti-A59V polyvalent ascitic fluid, (2) A59V MAb
AS, (3) A24, (4) A2 7 or (5) A 7 4 . Lanes 6-7 show BCVinfected cells with (6) anti-BCV polyvalent ascitic fluid
or (7) BCV MAb B24.
Lanes 8-13 show TGEV-infected cells
with (8) anti-TGEV polyvalent ascitic fluid, (9) TGEV MAb
T 4 1 , (10) T22, (11) T 20 , (12) T39 or (13) T40.
52
the other two recognized the nucleocapsid protein (lanes 3
and 4).
protein,
Several of the MAbs precipitated more than one
but none of the six cross-reacted with the
antigenicalIy related BCV.
Three of the anti-peplomer
antibodies neutralized the infectivity of A59V, but not of
BCV.
The other anti-P M A b , A74, had no detectable
neutralizing activity against A59V or BCV.
Only one
stable hybridoma line was isolated which produced
antibodies against BCV.
This MAb (B24) precipitated the
nucleocapsid proteins of both BCV and A59V.
The four
anti-CCV MAbs were all directed against the nucleocapsid
protein, and all four cross-reacted with TGEV.
Six MAbs
were isolated which recognized the 50 kd nucleocapsid
protein of TGEV.
Five of these cross-reacted strongly
with CCV by immunofluorescence.
Only one hybridoma cell
line produced antibodies against the TGEV pepIomer
glycoprotein, MAb T22 (Figure 11,
lane 10).
These
antibodies also cross-reacted with CCV and had
neutralizing activity against both TGEV and CCV.
None of
the antibodies directed against the nucleocapsid protein
showed neutralizing activity.
Most of the anti-N immuno-
precipitations showed a high molecular weight band which
probably represents aggregation or complexing of smaller
polypeptides, since it did not co-migrate with any of the
intracellular viral proteins.
N o monoclonal antibodies
recognizing the membrane glycoprotein(s) were isolated.
53
Cross-reactions of the monoclonal antibodies were
assessed by immunofluorescence.
Positive reactions were
verified by iirttnunoprecipitation.
Cross-reacting MAbs
precipitated homologous polypeptides from infected cell
lysates of the related virus.
shown in Figure 12A.
The results for TGEV are
The four anti-N MAbs of CCV
precipitated various combinations of intracellular
proteins from TGEV-infected cells.
The TGEV hybridoma T6
was the only one isolated against CCV or TGEV which failed
to cross-react with the related virus.
It precipitated
the 50 kd N protein and 4 other intracellular polypeptides
from TGEV-inf ected cell
lysates (Figure 12B,
none from CCV-infected cells (lane 3).
lane I),, but
TGEV MAb T7
reacted with the 50 kd proteins of TGEV and C C V , and the
40 kd
protein of TGEV
(lanes 2 and 4).
54
B
A
2
%
3
»»
I
#
M
50 »
s
Figure 12. Cross-immunoDrecipitation of viral proteins by
monoclonal antibodies.
^ S -methionine labeled
polypeptides from TGEV or CCV-infected cells were
immunoprecipitated with heterologous monoclonal antibodies
as described in Materials and Methods and the polypeptides
separated by SDS-PAGE in 8% slab gels.
Viral proteins are
indicated with their sizes in kd. Figure 12A shows TGEVinfected cell lysates immunoprecipitated with (I) anti-CCV
polyvalent ascitic fluid, (2) CCV-MAb C45, (3) C39, (4)
C115, or (5) C49.
Figure 12B shows TGEV-infected cell
lysates immunopre ci pi tated with (I) TGEV MAb T6 or (2) T 7 ,
and CCV-infected cell lysates immunoprecipitated with (3)
TGEV MAb T 6 or (4 ) T7.
55
DISCUSSION
This project had two objectives.
These were to
develop a reproducible laboratory method for comparing
antigens of diverse virus species, and to use the system
in a comprehensive investigation of the molecular
interrelationships among mammalian coronaviruses. This was
accomplished by preparing murine polyvalent and monoclonal
antibodies against four coronaviruses from two different
antigenic groups, and using them to compare the
intracellular proteins of the viruses.
Four different species of coronavirus were chosen for
study:
murine hepatitis virus A S 9 (A59V), bovine
coronavirus (BCV), canine coronavirus (CCV) and porcine
transmissible gastroenteritis virus (TGEV).
Cell culture '
systems were established and optimized for each virus.
The use of continuous cell lines made serial passage of
virus and maintenance of cells relatively simple.
Propagation of virus in cells with marked cytopathic
effect (CPE) allowed cloning by plaque selection and
quantitation of virus by plaque titration.
Previous studies have placed most of the known
mammalian coronaviruses into one of two antigenic groups
based on serological cross-reactions (reviewed in 68).
56
Most of these studies were conducted using heterotypic
serum, hnd'there was considerable ambiguity in the
results.
The use of heterotypic serum in interspecies
neutralization tests gives misleading results because of
I
endogenous antibodies directed against non-viral antigens
in the sample.
Antibodies which neutralized the infect-
ivity of TGEV in a complement-dependent reaction were
detected in serum from a number of animal species (19).
It was suggested that the neutralization was due to
heterophilic antibodies directed against porcine
glycolipids in the viral envelope.
The widespread
occurrence of BGV-neutraIizing antibodies in "normal"
human and animal sera (21,53) precludes the use of such
samples in studies of antigenic interrelationships.
These
antibodies could be the result of infection with BCV or a
related coronavirus, or the cross-neutralization may be
due to heterophilic antibodies such as those described
above.
I used anti-viral ascitic fluids prepared in
inbred mice as a homogeneous source of antibody against
A59V, B C V , CCV and TGEV.
Infected-cel I lysates were used ■
as immunogen, eliminating the need for virus purification.
Specific high-titer antibody was obtained in as little as
four weeks.
Large volumes of fluid accumulated in
individual mice, so that a single immunization schedule
yielded enough antibody for a complete battery of
serological tests.
The ascitic fluids detected the
57
intracellular polypeptides of the immunizing viruses as
well as the structural proteins (Figure 5).
The titers of
the fluids compared favorably to serum in homologous virus
neutralization tests (Figure 3).
Although immunization
with cell lysates produced antibodies against cellular as
well as viral proteins, the non-viral antibodies were
successfully removed by absorbing the ascitic fluids with
uninfected cells
(Figure 2).
Monoclonal antibodies
against al I four viruses were produced by fusion of
splenocytes from the immunized mice with myeloma cells,
providing a bank of specific reagents for precise
investigation of viral antigens at the level of the
individual polypeptides.
Thus this protocol fulfilled
each of the criteria set down for e s t a b l i s h m e n t of a
standardized system for studying antigenic
interrelationships among viruses of different species.
Serological Comparisons
The absorbed hyperimmune ascitic fluids were used to
investigate the antigenic interrelationships among the
four coronaviruses.
The IFA confirmed the cross-reactions
described previously between A59V and BCV (26) and between
CCV and TGEV (18,29 ,49).
No cross-reactions were seen
among viruses from different antigenic groups
(Figure 6).
Reciprocal VN tests showed the same relationships.
58
although the data seemed to suggest a closer relationship
between CC V and TGEV than between A S 9V and BCV (Table 3).
Molecular Comparisons
Reciprocal cross-immunoprecipitations of intracellular
viral proteins showed that the cross-reactions between
related coronaviruses were due to common antigenic
determinants on each of the major viral antigens:
the
p e p lomer, nucleocapsid and membrane proteins (Figures 710).
The immunoprecipitation and monoclonal antibody
studies confirmed that TGEV and CCV are more closely
r e l a t e d than BCV and AS 9 V.
Of the ten anti-N MAbs
i s o l a t e d for TGEV and C C V , o n l y T 6 did not show c r o s s ­
reactivity with the related v i r u s , indicating that the N ■
proteins of CCV and TGEV are very closely related (Table
4, Figure 12B).
The anti-P MAb, T 2 2 , neutralized the
infectivity of both CCV and T G E V .
In contrast, none of
the A59V MAbs cross-reacted with BCV by IFA,
immunoprecipitation or VN.
Intracellular Virus-Specific Proteins'
The intracellular proteins of each virus were
identified by immunoprecipitation of radiolabeled
polypeptides from infected cell lysates with hyperimmune
ascitic fluid.
The intracellular proteins found in A59V-
cells were in general agreement with those reported in the
59
literature
(3,4,20,38,57,67).
The 26,
60 and 155 kd
proteins represent the major M , N and P antigens of the
virus.
The variable 45-57 kd proteins are probably
nucleocapsid degradation products (57).
However,
an
intracellular protein about 10 kd smaller than the N
protein was reproducibly precipitated from infected cell
lysates of all four viruses by polyvalent ascitic fluid
and anti-N monoclonal antibodies,
suggesting that this
protein may be an intracellular precursor of the N '
protein.
Non-structural proteins of B C V , CCV and TGEV have not
been described previously.
Four of the polypeptides
precipitated from BCV-infected cells had the same
approximate molecular weights as the structural proteins
reported by others (26,33,36).
Additional polypeptides of
26 and 50 kd may be processing intermediates, analogous to
the 18 and 50 kd intracellular proteins of A5 9V.
Maj o r p o l y p e p t i d e s of 185, 50 and 30 kd were
immunoprecipitated from lysates of CCV-infected A-72
cells.
These correspond to the three structural proteins
described by Garwes and Reynolds (18).
They also reported
the presence in virions of a 23 kd g l y c o p r o t e i n which I
did not find in infected cell lysates.
However,
the M
proteins of coronaviruses are known to be heterogeneous,
and the migration characteristics and number of bands seen
60
in SDS-PAGE vary widely in different laboratories
(3,4,57,65,67).
The intracellular proteins of TGEV proved to be quite
interesting.
In addition to the proteins homologous to ■
those described for CCV, three polypeptides of 77-92 kd '
were repeatedly precipitated from infected-cell
lysates.
An obvious hypothesis is that they represent variously
glycosylated monomers of the peplomeric protein,
since the
P proteins of A59V and BCV reportedly consist of dimers
(20,48), and .intracellular proteins of 90-100 kd were
often seen in gels of A59V, BC V and CCV.
However, the
recognition of these proteins by several of the anti-N
MAbs (Figure 12) suggests that they are related to the
50 kd nucleocapsid protein and not to the peplomeric
glycoprotein.
Monoclonal Antibodies
Most of the MAbs isolated were directed against
antigenic determinants on the n u c leocapsid (N) protein.
This was not unexpected, since this polypeptide is
synthesized in the greatest amounts in virus-infected
cells (57). Vautherot et al.
(73,74) used purified virions
to immunize mice for production of MAbs.
Most of their
antibodies had specificities for the peplomeric
glycoproteins,
presumably because these antigens are
exposed on the viral surface.
61
Several of the MAbs precipitated more than one viral
polypeptide.
This could be due to (i) failure of t h e ’
cloning procedure to separate hybridomas with different
specificities,
(ii) shared epitopes on the polypeptides
because of primary sequence homologies,
(iii) breakdown or'
aggregation of polypeptides to form antigens which migrate
differently in SDS-PAGE gels but maintain the same
specificity,
or (iv) non-specific binding of antibody or
S. aureus bacteria to viral proteins.
Al I MAbs were
cloned at least twice by limiting dilution,
and only
single colonies were selected for testing and subcloning.
Therefore the presence of mixed clones in the final
population is unlikely.
Also, several hybridomas were
recloned to maintain stability and showed no changes in
antibody specificity upon repeat testing.
Preparation and treatment of samples during
immunoprecipitation and gel electrophoresis can lead to
anomalous migration patterns of coronavirus glycoproteins
(65,67), giving multiple bands which in fact represent
only one protein.
I have done immunoprecipitations of
T G E V , CCV and M H V - infected cell lysates under a variety of
reducing and denaturing conditions.
There were no
substantial variations in the electrophoretic profiles
when sample treatments were changed.
Non-specific reactions can be a problem in immuno­
precipi tations using S. aureus. However, these are usually
62
unpredictable and irreproducible.
The intracellular
proteins precipitated by MAbs in these experiments were
seen in gels of two or more independent immunoprecipitations.
Also, control experiments including the immuno-
precipitation of infected-cell
lysates with normal mouse
ascitic fluid, heterologous non-reactive antibodies, and
S. aureus in the absence of antibody were done to detect
the presence of artifacts.
Therefore it seems likely that
the proteins precipitated by the MAbs are related because
of primary sequence homologies.
Conclusions
The antigens of four coronaviruses were compared using
several serological and molecular techniques.
The
serological comparisons were facilitated by the
development of a homogeneous system for production of
hyperimmune ascitic fluids in mice immunized with virusinfected cell lysates.
Intracellular proteins of BCV, CCV
and T G E V , which have not been described previously, were
identified using these anti-viral ascitic fluids.
I
Monoclonal antibodies were isolated against
nucleocapsid and peplomeric proteins,of AS9V and TG E V , and
against the N proteins of BCV and C C V . These MAbs
provided a powerful tool for analyzing antigenic
interrelationships among viruses and for studying the
biological activity of viral polypeptides (e.g. eliciting
63
of neutralizing antibodies by the peplomeric
glycoprotein).
MAbs are also valuable for investigation
of intracellular protein processing events taking place
during virus infection, and determining the relationships
between intracellular and structural proteins.
The coronaviruses TGEV and CCV were found to be very
closely related, with common epitopes on all major viral '
antigens.
Although no MAbs were isolated which recognized
the M proteins, polyvalent anti-viral ascitic fluids
precipitated this protein in reciprocal crossimmunoprecipitation tests.
Only one of eleven MAbs failed
to show cross-reaction between CCV and TGEV.
This
hybridoma could be a useful diagnostic reagent for
differentiation of the two viruses in clinical samples.
In conclusion, this research has introduced a
standardized and comprehensive system for studying viral
antigens and has illustrated the applications of this
method in increasing the knowledge of coronavirus
interrelationships.
64
LITERATURE CITED
65
1.
B i n n , L.N., E.C. Lazar, K.P. Keenan, D.L. Huxsol I ,
R.H. Marchwicki and A. Strano. 1974. Recovery and
characterization of a coronavirus from military dogs
with diarrhea. Proc. Ann. Meet. U.S. Anim. Health
Assoc. 78: 359-366.
2.
Binn, L.N., R.H. M a r c h w i c k i and E.H. Stephenson.
1980. Establishment of a canine cell line:
derivation, characterization, and viral spectrum. Am.
J. Vet. Res. 4JL: 855-860.
3.
Bond, C.W., K.Anderson and J.L. Leibowitz. 1984.
Protein synthesis in cells infected by murine
hepatitis viruses JHM and A59:
tryptic peptide
analysis. Arch. Virol. 8_0 : 333-347.
4.
Bond, C.W., K. Anderson, S . Goss, and L. Sardinia.
1981. Relatedness of virion and intracellular
proteins of the murine coronaviruses JHM and A59.
A d v. Exp. Med. Biol. 14 2 : 103-110.
5.
B u r k s , J.S., B.L. DeVa I d, L.D. Janko v s k y and J.C.
Gerdes. 1980. Two coronaviruses isolated from central
nervous system tissue of two multiple sclerosis
patients. Science 209: 933-934.
6.
Bonner, W.M. and R.A. Laskey, 19 7 4. A fil m detection
method for tritium-IabeI led proteins and nucleic
acids in polyacrylamide gels. Eur. J. Biochem. 46:8388.
7.
Bradburne, A.F. 197 0. Antigenic relationships amongst
corona viruses. Arch. Virol. 3_1: 352-364.
8.
Cheley, S., V.L. Morris, M.J. Guppies, and R.
Anderson. 1981. RNA and polypeptide homology among
coronaviruses. Virology 115: 310-321.
9.
Childs, J.C., S.A. Stoh I m a n , L. Kingsf or d , and R.
Russell. 1983. Antigenic relationships of murine
coronaviruses. Arch. Virol. TjB: 81-87.
10.
Collins, A.R., R.L. Knob I e r , H. Powell, and M.J.
Buchmeier. 1982. Monoclonal antibodies to murine
hepatitis virus-4 (strain J H M ) define the viral
glycoprotein responsible for attachment and cell-cell
fusion. Virology 119: 358-371.
66
11.
D e a , S., R.S. Roy and M.A.S.Y. Elazhary. 1 982.
Antigenic variations among calf diarrhea
coronaviruses by immunodiffusion and
counterImmunoelectrophoresis. Ann. Rech. Vet. 13:
351-356.
—
12.
Debouck, P. and. M. Pensaert. 19 80. Experimental
infection of pigs with a new porcine enteric
coronavirus, CV 777. Am. J. Vet. Res. 4:1: 219-223.
13.
de St. Grothf S.F. and D. S c h e i d e g g e r . 1980.
Production of monoclonal antibodies: Strategy and
tactics. J. Immunol. Methods 3J5: 1-21.
14.
Doyle, L.P. and L.M. Hutchings. 1946. A transmissible
gastroe n t e r i t i s in pigs. J. Am. Vet. Assoc. 108: 25 7259.
15.
Fairbanks, G., T.L. Stock, and D.F.H. W a l l ach. 1971.
Electrophoretic analysis of the major polypeptides of
the human erythrocyte membrane. Biochemistry 10:
2606-2617.
16.
Fleming, J.O., S.A. S t o h l m a n , R.C. Harmon, M.M.C.
Lai, J.A. F r e I i n g e r , and L.P. Weiner. 1983. Antigenic
relationships of murine coronaviruses: Analysis using
monoclonal antibodies to JHM (MHV_4) virus. Virology
131: 296-307.
17.
G a r w e s , D.J. and D.H. Pocock. 1975. The p o l y p e p t i d e
structure of transmissible gastroenteritis virus. J.
Gen. Virol. 2_9: 25-34.
18.
Garwes, D.J. and D.J. Reynolds. 1981. The polypeptide
structure of canine coronavirus and its relationship
to porcine transmissible gastroenteritis virus. J.
Gen. Virol. 52^ 153-157.
19.
G a r w e s , D. J., M.H. Lucas, D.A. Higgins, B.V. P i k e ,
and S.F. Cartwright. 1978. Antigenicity of structural
components from porcine transmissible gastroenteritis
virus. Vet. Microbiol. _3: 179-190.
20.
G e r d e s , J.C., I. Klein, B.L. D e V a l d and J. S . B u r k s .
1981. Coronavirus isolates SK and SD from multiple
sclerosis patients are serologically related to
murine coronaviruses A59 and JHM and human
coronavirus OC43, but not to human coronavirus 229E.
J. Virol . 38: 231-238.
67
21.
Gerna, G. f P.M. C e r e d a , M.G. R e v e l I o, E . Cattaneo, M.
Battaglia and M. Torse 11 ini-Gerna. 1981. Antigenic
and biological relationships between human
coronavirus- OC43 and neonatal calf diarrhea virus, J.
Gen. Virol. 5_4: 91-102..
22.
Hajer, I. and J. S t o r z . 1978. Antigens of bovine
coronavirus strain LY-138 and their diagnostic
properties. Am. J. Vet. Res. 3_9: 441-444.
23.
Hajer, I. and J. Storz. 1979. Structural polypeptides
of the enteropathogenic bovine coronavirus strain LY138. Arch. Virol. 5_9: 4 7-5 7.
24.
Hasony, H.J. and M.R. MacNaughton. 1981. Antigenicity
of mouse hepatitis virus strain 3 subcomponents in
C57 strain mice. Arch. Virol. 6_9: 33-41.
25.
Herrmann, E.C.Jr. and Claire Engle. 1958. Tumor cellinduced mouse ascities fluid as a source of viral
antibodies. Proc. Soc. Exp. Biol. Med. 9_8 : 257-259.
26.
Hogue, B.G., King, B. and D.A. Brian. 1984. Antigenic
relationships among proteins of bovine coronavirus,
human respiratory coronavirus OC43, and mouse
hepatitis coronavirus A59. J. Virol. _51: 384-388.
27.
Holmes. K.V., Boiler, E.W. and J.N. Behnke. 1981.
Analysis of the functions of coronavirus
glycoproteins by differential inhibition of synthesis
with tunicamycin. Adv. Exp. Med. Biol. 142; 133-142.
28.
Holmes, K.V., E.W. Boiler, and L.S. Sturman. 19 81.
Tunicamycin resistant glycosylation of a coronavirus
glycoprotein: Bemonstration of a novel type of viral
glycoprotein. Virology 115: 334-344.
29.
H o r zinek, M.C., H. Lutz, and N.C. Pederson. 19 83.
Antigenic relationships among homologous structural
polypeptides of porcine, feline, and canine
corona viruses. Infect. Immun. 3J7: 1148-1 15 5.
30.
Kaye, H.S., W.B. Yarbrough, C.J. Reed and A.K.
Harrison. 1977. Antigenic relationship between human
coronavirus strain OC-43 and hemagglutinating
encephalomyelitis virus strain 67N of swine: antibody
responses in human and animal sera. J. Infect. Bis.
135: 201-209.
68
31.
K e e n a n , K.P., H.R. Jervis, R .H . March w i c k i and L.N.
Binn. 1976. Intestinal infection of neonatal dogs
with canine coronavirus 1-71: studies by virologic,
histologic, histochemicaI, and immunofluorescent
techniques. Am. J. Vet. Res. JBTg 247-256.
32.
Kessler, S.W. 1975. Rapid isolation of antigens from
cells with a staphylococcal protein A-antibody
adsorbent: Parameters of the interaction of antibodyantigen complexes with protein A. J. Immunol: 115:
1617-1624.
33.
King, B. and D.A. Brian. 1982. Bovine coronavirus
structural proteins. J. Virol. 42g 7 00-7 07.
34.
Laemmli, U.K. and M. Favre. 1973. Maturation of the
head of bacteriophage T4. I. DNA packaging events. J.
Mol. Biol. 80: 575-599.
35.
Laport e, J., P. Bobu I esco, and F . Rossi. 1980. Une
lignee cel lulaire particulierement sensible a la
replication du Coronavirus enterique bovin: Ies
c e l l u l e s HRT 18. C. R. Acad. Sci. Paris. 290 : 623626 .
36.
Laporte, J. and P. Bob u lesco. 1981. Polypeptide
structure of bovine enteritic coronavirus: comparison
betw e e n a w i l d strain purified from feces and a HRT18 adapted strain. A d v . Exp. Med. Biol. 1 4 2 : 181-184.
37.
MacNaughton, M.R., M.H. M a d g e and S.E. Reed. 19 81.
Two antigenic groups of human coronaviruses detected
by using enzyme-Iinked immunosorbent assay. Inf.
Immun. 3_3 : 734-7 37.
38.
M a h y , Brian W.J. 1981. Biochemistry of coronaviruses
1980. A d v . Exp. Med. Biol. 14 2 : 261-270.
39.
M c C Iurkin, A.W. and J.O. Norman. 1966. Studies on
transmissible gastroenteritis of swine. II. Selected
characteristics of a cytopathogenic virus common to
five isolates from transmissible gastroenterits. Can.
J. Comp. Med. Vet. 3J): 19 0-198.
40.
McIntosh, K., A.Z. K a p i k i a n , K.A. Hardison, J.W.
Hartley and R.M. Chanock. 1969. Antigenic
relationships among coronaviruses of man and between
human and animal coronaviruses. J. Immunol. 102:
1109-1118.
69
41.
McIntosh, K. 1.974. Corona viruses: A comparative
review. Curr. Topics Microbiol. Immunol. ^3: 85-129.
42.
M e b u s , C.A., E.L. Stair, M.B. Rhodes, and M.J.
Twiehaus. 1973. Neonatal calf diarrhea: propagation,
attenuation, and characteristics of a coronavirus­
like agent. Am. J. Vet. Res. 3_4: 145-150.
43.
Pederson, N.C., J. W ard and W.L. Menge I ing. 19 78.
Antigenic relationship of the feline infectious
peritonitis virus to coronaviruses of- other species.
Arch. Virol. 5_8: 45-53.
44.
P e n s a e r t , M.B. and P. Debouck. 19 7 8. A new
coronavirus-I ike particle associated with diarrhea in
swine. Arch. Virol. 5_8: 243-247.
45.
Pens a e r t , M.B., P. Debouck and D.J. Reynolds. 19 81.
An immunoelectron microscopic and immunofluorescent
study on the antigenic relationship between the
coronavirus-like agent, CV-777, and several
corona viruses. Arch. Virol. 6_8: 45-52.
46.
Pike, B.V. and D.J. Garwes. 1979. The neutralization
of transmissible gastroenteritis virus by normal
heterotypic serum. J. Gen. Virol. 4_2: 279-287.
47.
Reynolds, D.J., D.J. Garwes and G.L. Gaskel I. 1977.
Detection of transmissible gastroenteritis virus
neutralising antibody in cats. Arch. Virol. _55: 7786.
48.
Reynolds, D.J. and D.J. Garwes. 1979. Virus isolation
and serum antibody responses after infection of cats
with transmissible gastroenteritis virus. Arch.
Virol. 6_0: 161-166.
49.
R e y n o l d s , O.J., D.J. Garwes and S. L u c e y . 19 80.
Differentiation of canine coronavirus and porcine
transmissible gastroenteritis virus by neutralization
with canine, porcine and feline sera. Vet. Microbiol.
- 5: 283-290.
50.
Robb, James A. 1973. Microculture Procedures. B.
Simian Virus 40. In: Tissue Culture Methods and
Applications, P.F. Kruse and M.K. Patterson, eds.
Academic Press, New York.
51.
Robb, J.A. and C.W. Bond. 1979. C o r o n a v i r i d a e .
Comprehensive Virol. L 4 : 193-247.
70
52.
S a t o , K . , Y. Inaba and M. Matumoto. 1980. Sero logic
relationship between calf diarrhea coronavirus and
hemagglutinating encephalomyelitis virus. Arch.
Virol. 66: 157-159.
53.
Sato, K., Y. Inaba, Y. Miura, S. T o k u h i s a , H. A k a s h i ,
T. Shinosaki and M. Matumoto. 1981. Neutralizing
antibody to calf diarrhea coronavirus in various
animal species in Japan.
Microbiol. Immunol. 25:
623-625.
54.
Schmidt, O.W. and G.E. Kenny. 1981. Immunogenicity
and antigenicity of human coronaviruses 229E and OC43. Inf. Imm. 3_2 : 1000-1006.
55.
Schmidt, O.W. and G.E. Kenny. 1982. P o l y p e p t i d e s and
function of antigens from human coronaviruses 229E
and 0 C 4 3. Infect. Immun. 3_5: 515-522.
56.
Shapiro, A.L., E. Vihue I a, and J.V. M a i z e l . 1967.
Molecular weight estimation of polypeptide chains by
electrophoresis in SDS-pol yacry I amide gels. Bioc hem.
Biophys. Res. Comm. 2_8: 815-820.
57.
Siddel I, S.G., H. Wege, and V. ter Meulen. 1982. The
structure and replication of coronaviruses. Curr.
Topics in Microbiol. Immunol. 9_9: 131-163.
58.
Siddel I, S., H. Wege, A. B a r t h e l , and V. ter Meulen.
1981. Coronavirus J H M : Intracellular protein
synthesis. J. Gen. Virol. 5J3 : 145-155.
59.
Small, J.D., L. Aurelian, R.A. Squire, J.D.
S t r a n d b e r g , E.C. Melby, Jr., T.B. Turner
and B. Newman. 1979. Rabbit cardiomyopathy
associated with a virus antigenicaI Iy
related to human coronavirus 229E. Am. J.
Pathol. 95: 709-724.
60.
Stair, E.L., M.B. Rhodes, R.G. White, and C.A. M e b u s .
1972. Neonatal calf diarrhea: Purification and
electron microscopy of a coronavirus-1ike agent. Am.
J. Vet. Res. 33: 1147-1156.
61.
Storz, J. and R. Rott. 1981. Reactivity of antibodies
in human serum with antigens of an enteropathogenic
bovine coronavirus. Med. Microbiol. Immunol. 169:
169-178.
71
62.
S t o r z , J., G. Kaluza, H. Niemann and R. R o t t . 19 81.
On enteropathogenic bovine coronavirus. Adv. Exp.
Med. Biol. 142: 171-179.
63.
Sturman, L.S. and K.K. T a k e m o t o . 1972. Enhanced
growth of a murine coronavirus in transformed mouse
cells.
Infect. Immu n. 6^ 501-507.
64.
S t u r m a n , L.S., K.V. Holmes and J. Behnke. 1980.
Isolation of coronavirus envelope glycoproteins and
interaction with the viral nucleocapsid. J. Virol.
3_3: 449-462.
65.
Sturman, L.S. 1977. Characterization of a
coronavirus. I. Structural proteins: effects of
preparative conditions on the migration of protein in
polyacrylamide gels. Virology 7J7: 637-649.
66.
Sturman, L.S. and K.V. Holmes. 1977. Characterization
of a coronavirus. II. Glycoproteins of the viral
envelope: tryptic peptide analysis. Virology 7_7: 6506 60.
67.
Sturman L.S. 1981. The structure and behavior of
• coronavirus A S 9 glycoproteins. Adv. Exp. Med. Biol.
142: 1-17.
68.
Sturman, L.S. and K.V. Holmes. 1983. The molecular
biology of corona viruses ^ Adv. Virus Res. 2_8: 35-112.
69.
Tompkins, W.A.F., A.M. Watrach, J.D. Schmale, R.M.
Schultz and J.A. Harris. 1974. Cultural and antigenic
properties of newly established cell strains derived
from adenocarcinomas of the human colon and rectum.
J. Natl. Cancer Inst. 5^2: 1101-1110.
70.
Tikasingh, E.S., L.
use of adjuvant and
production of mouse
arboviruses. Am. J.
71.
Tung, Amar S. 1983. Production of large amounts of
antibodies, non-specific immunoglobulins and other
serum proteins in ascitic fluids of individual mice
and guinea pigs. Meth. Enzymol. 9_3: 2-23.
72.
Tyrrell, D.A.J., J.D. Almeida, D .M . Berry, C.H.
Cunningham, D. Harare, M.S. Hofstad, L. Mallucci and
K. McIntosh. 1968. Coronaviruses. Nature (Bond.) 220:
650:
Spence, and W.G. Downs. 1966. The
sarcoma 180 cells in the
hyperimmune ascitic fluid to
T r o p . Med. Hyg. _1_5: 219-226.
72
73.
Vautherot, J-F. 1981. Plaque assay for titration of
bovine enteric coronavirus. J. Gen. Virol. 56- 451455.
—
74.
Vautherot, J.F., J. Laporte, M.F. Madelaine, P.
Bobulesco and A. Roseto. 1981. Antigenic and
polypeptide structure of bovine enteric coronavirus
as defined by monoclonal antibodies. Adv. Exp. Med.
Biol. 173: 117-132.
75.
Vautherot, J.F. and J. Laporte. 1983. Utilization of
monoclonal antibodies for antigenic characterization
of coronaviruses. Ann. Rech. Vet. JMl: 437-444. '
76.
Wege, H., Siddel I, and V. ter Meulen. 1982. The
biology and pathogenesis of coronaviruses. Curr.
Topics Microbiol. Immunol. 9_9: 165-200.
77.
Woods, R.D. and M.C. Pedersen. 1979. Cross-protection
studies between feline infectious peritonitis and
porcine transmissible gastroenteritis viruses. Vet.
Microbiol. _4: 11-16.
78.
Woods, R.D. 1982. Studies of enteric coronaviruses in
a feline cell line. Vet. Microbiol. J7: 427-435.
79.
Yaseen, S.A. and
Antigenic studies
of the structural
strain 229E, Can.
C.M. Johnson-Lussenburg. 1981.
on coronaviruses. I. Identification
antigens of human coronavirus,
J. Microbiol. 27: 334-342.
MONTANA STATE UNIVERSITY LIBRARIES
3
762 1001 816 4
D378
G69k
cop. 2
GkD s s , s. H.
Serological and mo l e ­
cular c o m p a r i s o n s ...
DATE
D378
G69U
cop.2