JOURNAL OF VIROLOGY, Feb. 1992, p. 956-965 Vol. 66, No.

2
0022-538X/92/020956-10$02.00/0
Copyright X 1992, American Society for Microbiology

Monoclonal Antibodies to the Spike Protein of Feline Infectious

Peritonitis Virus Mediate Antibody-Dependent Enhancement
of Infection of Feline Macrophages
CHRISTOPHER W. OLSEN,* WAYNE V. CORAPI, CHRISTOPHER K. NGICHABE,
JOEL D. BAINES,t AND FRED W. SCOTT
Department of Microbiology, Immunology, and Parasitology and Cornell Feline Health Center, New York
State College of Veterinary Medicine, Cornell University, Ithaca, New York 14853
Received 15 August 1991/Accepted 7 November 1991

Antibody-dependent enhancement of virus infection is a process whereby virus-antibody complexes initiate

infection of cells via Fc receptor-mediated endocytosis. We sought to investigate antibody-dependent enhance-
ment of feline infectious peritonitis virus infection of primary feline peritoneal macrophages in vitro.
Enhancement of infection was assessed, after indirect immunofluorescent-antibody labelling of infected cells,
by determining the ratio between the number of cells infected in the presence and absence of virus-specific
antibody. Infection enhancement was initially demonstrated by using heat-inactivated, virus-specific feline
antiserum. Functional compatibility between murine immunoglobulin molecules and feline Fc receptors was
demonstrated by using murine anti-sheep erythrocyte serum and an antibody-coated sheep erythrocyte
phagocytosis assay. Thirty-seven murine monoclonal antibodies specific for the nucleocapsid, membrane, or
spike proteins of feline infectious peritonitis virus or transmissible gastroenteritis virus were assayed for their
ability to enhance the infectivity of feline infectious peritonitis virus. Infection enhancement was mediated by
a subset of spike protein-specific monoclonal antibodies. A distinct correlation was seen between the ability of
a monoclonal antibody to cause virus neutralization in a routine cell culture neutralization assay and its ability
to mediate infection enhancement of macrophages. Infection enhancement was shown to be Fc receptor
mediated by blockade of antibody-Fc receptor interaction using staphylococcal protein A. Our results are
consistent with the hypothesis that antibody-dependent enhancement of feline infectious peritonitis virus
infectivity is mediated by antibody directed against specific sites on the spike protein.

Feline infectious peritonitis virus (FIPV) is a member of new consideration was introduced in 1989 when Stoddart
the family Coronaviridae. Coronaviruses are plus-sense, provided evidence for antibody-dependent enhancement
single-stranded RNA viruses with three major structural (ADE) of FIPV infection of primary feline peritoneal mac-
proteins, the spike (S), membrane (M), and nucleocapsid (N) rophages in vitro (41).
proteins (18). FIPV infects domestic as well as exotic cats ADE of virus infection occurs when monocytes and mac-
and produces an ultimately fatal disease called feline infec- rophages are more efficiently infected by complexes of virus
tious peritonitis (FIP) (2, 30, 39). One of the most perplexing plus antibody (Ab), via Fc receptor-mediated endocytosis,

Downloaded from https://journals.asm.org/journal/jvi on 02 January 2025 by 45.78.52.78.

aspects of the pathogenesis of FIP is the frequent occurrence than by virus alone (33). A number of human and animal
of accelerated, more fulminant disease upon FIPV challenge viruses have been shown to be capable of utilizing this
of seropositive as compared with seronegative cats. Accel- mechanism of infection. These include dengue virus and
erated FIP was first documented when Pedersen and Boyle related flaviviruses (8, 32), respiratory syncytial virus (5, 17),
showed that the onset of clinical disease among experimen- influenza virus type A (26, 27), rabies virus (16), and most
tally infected kittens correlated with the appearance of recently human immunodeficiency virus type 1 (11, 44), as
serum antibodies (31). Weiss and Scott confirmed these well as FIPV (41). The FIPV system provides a unique
results, demonstrating that the onset of viremia, clinical opportunity to investigate ADE of virus infectivity at both
signs, thrombocytopenia, lymphopenia, and the appearance the in vitro and in vivo levels. Specific aspects of ADE of
of viral antigen and necrotizing lesions in affected tissues all FIPV infection need to be elucidated because the immuno-
occurred earlier in seropositive kittens than in seronegative pathogenesis of FIP has precluded the development of an
kittens (48, 49). Survival times were also significantly shorter effective, proven vaccine against FIPV infection (28).
for seropositive kittens (31, 48). In addition, seronegative The aims of this work were to confirm that virus-specific
kittens given immune serum (31, 47) or anti-FIPV immuno- feline antiserum can mediate ADE of FIPV infectivity in
globulin G (IgG) (31) before challenge developed clinical vitro and to determine whether FIPV-specific murine mono-
disease in the same manner and over the same time course as clonal Abs (MAbs) can demonstrate similar in vitro enhance-
did seropositive kittens. The demonstration of immune com- ment. A panel of MAbs was evaluated to define which viral
plex deposition in FIP by Jacobse-Geels et al. (13, 14) protein(s) induces enhancing Abs and to determine whether
initially seemed to explain both the pathologic changes of distinct enhancing epitopes may be involved.
FIP and the phenomenon of accelerated FIP. However, a
MATERIALS AND METHODS
*
Corresponding author.
t Present address: Marjorie Kovler Oncology Laboratories, Uni- Cells and viruses. FIPV strains 79-1146 and UCD1 were
versity of Chicago, Chicago, IL 60637. grown in A72 cells. Virus titers were calculated as 50%
956
VOL. 66, 1992 ANTIBODY-DEPENDENT ENHANCEMENT OF FIPV 957

tissue culture infectious dose (TCID50) units by the method oratories, Grand Island, N.Y.) per ml. Macrophages were
of Reed and Muench (34), using Crandell feline kidney cells concentrated by centrifugation over a 62% Percoll (Pharma-
(CrFKC) as indicator cells. cia Fine Chemicals, Piscataway, N.J.) cushion and then
Polyclonal Ab preparations. Murine anti-sheep erythrocyte pelleted, washed, and resuspended in complete macrophage
(anti-SRBC) serum was obtained from a BALB/c mouse medium (CMM; L-15 [GIBCO], 20% heat-inactivated, char-
(The Jackson Laboratory, Bar Harbor, Maine) following a acterized fetal bovine serum [Hyclone Laboratories, Logan,
series of three intraperitoneal inoculations of 0.5 ml of a 50% Utah], 4 mM added glutamine [GIBCO], 20 p,g of gentamicin
suspension of freshly collected, washed SRBC. The serum [GIBCO] per ml) to a concentration of 7.5 x 105 cells per ml.
was heat inactivated by heating to 56°C for 30 min to remove Macrophages were cultured on eight-chamber glass Lab-Tek
complement activity. slides (Nunc, Inc., Naperville, Ill.) at a concentration of 1.5
FIPV-specific feline antiserum was kindly provided by x 105 cells per chamber. Macrophage cultures were main-
Cheryl Stoddart (Stanford University, Palo Alto, Calif.). tained at 37°C in the absence of CO2. Cell chambers were
MAbs. MAbs to FIPV were produced by using standard washed 7 h postseeding with PBS to remove nonadherent
techniques (3). Briefly, a semipurified, whole virus prepara- cells, after which fresh medium was added. Adherent cells
tion of FIPV 79-1146 was used to immunize BALB/c mice were characterized as macrophages on the basis of morpho-
(Charles River Breeding Laboratories, Inc., Wilmington, logic appearance, demonstration of nonspecific esterase
Mass.). Mice were immunized three times and boosted with activity (Sigma kit 90; Sigma Chemical Co., St. Louis, Mo.),
an intrasplenic injection 3 days prior to fusion. Hybridoma and demonstration of functional Fc receptors in an Ab-
colonies were cloned by limiting dilution, and viral speci- coated SRBC phagocytosis assay (see below).
ficity was assessed by indirect immunofluorescent-Ab (IFA) Ab-coated SRBC phagocytosis assay. An Fc receptor-me-
labelling of infected CrFKC. Virus protein specificity was diated phagocytosis assay was conducted as previously
evaluated by radioimmunoprecipitation assay (RIPA) (3), described (42). Fresh SRBC were pelleted and washed three
and IgG subclass was determined by using a commercially times and then resuspended to 5% (vol/vol) in PBS. Equal
available kit (Zymed Laboratories, Inc., South San Fran- volumes of this 5% SRBC solution and either rabbit anti-
cisco, Calif.). SRBC serum (1:128) (Organon Teknika, Durham, N.C.),
Two MAbs raised against the DF2 strain of FIPV and two murine anti-SRBC serum (1:128) (see above), or PBS were
MAbs raised against the Miller strain of transmissible gas- incubated at 37°C for 30 min. The erythrocyte-Ab complexes
troenteritis virus were kindly supplied by Susan Fiscus were then pelleted and washed three times in PBS and
(University of North Carolina, Chapel Hill) and Fermenta resuspended to 1% (vol/vol) in L-15 medium. Macrophages
Animal Health. The viral protein specificity and immuno- in eight-chamber Lab-Tek slides were inoculated with the
globulin subclass of these MAbs have been previously 1% SRBC solutions (0.1 ml per chamber) and incubated for
reported (4). 60 min at 37°C. The slides were then rinsed in PBS, dipped
All of the Ab preparations (sera and ascites fluid) for the in distilled water, and stained with May-Grunwald-Giemsa
enhancement assays were heat inactivated prior to use. This stain.
was done in order to exclude complement-mediated en- In vitro assay for enhancement of FIPV infectivity. The
hancement of infection, which has been documented in other assay for enhancement of virus infectivity was adapted from
virus systems (33). the method described by Stoddart (41). Macrophages were
Virus neutralization assay. Each of the MAbs was assayed infected with virus 22 to 24 h after having been seeded onto
for virus neutralization activity on CrFKC by using a 96-well Lab-Tek slides. Each MAb (as ascites fluid) or antiserum
plate format. Serial twofold dilutions of heat-inactivated sample was initially evaluated at a dilution (in CMM) of 1:64
MAb were mixed with an equal volume of either FIPV and serial twofold dilutions from 1:512 to 1:16,384. FIPV
79-1146 or FIPV UCD1 containing 100 TCID50 of virus. UCD1 stock virus was diluted to 2 x 106 TCID5Jml in

Downloaded from https://journals.asm.org/journal/jvi on 02 January 2025 by 45.78.52.78.

Following a 1-h incubation at room temperature, CrFKC CMM. Equal volumes of diluted virus plus Ab (or plus CMM
were added and plates were incubated at 37°C for 96 h. Titers for the wells to be infected by virus alone) were mixed and
are reported as the reciprocal of the highest dilution of Ab incubated at 37°C for 30 min. Macrophage cultures in tripli-
that completely inhibited cytopathology. cate wells were then washed once with PBS and inoculated
Coliection and purification of primary feline peritoneal with 400 ,ul of preincubated virus plus Ab or virus plus
macrophages. Macrophages were obtained by saline lavage medium. Inoculation with CMM alone and mixtures of MAb
of the peritoneal cavity (42) of specific-pathogen-free (SPF) or antiserum plus medium served as negative controls. To
cats (Liberty Laboratories, Liberty Corners, N.J.). Before rule out enhancement of virus infection by nonspecific serum
use as a macrophage donor, each cat was confirmed to be factors, heat-inactivated preimmune feline serum as well as
coronavirus antibody negative by kinetics-based enzyme- coronavirus-negative murine serum was incubated with
linked immunosorbent assay at the Diagnostic Laboratory at FIPV as described above, and the level of infection achieved
the New York State College of Veterinary Medicine at was compared with that obtained with virus alone.
Cornell University and by serum neutralization testing. The Following inoculation, macrophages were cultured for 60
peritoneal lavage technique could be performed on a cat min at 37°C, after which the Lab-Tek chambers were thor-
repeatedly at 9- to 12-day intervals without discomfort or oughly washed with PBS to remove nonadsorbed virus and
other side effects, thus reducing the number of laboratory replenished with 300 ,ul of CMM. At 10.5 h after infection,
animals required. the Lab-Tek slides were disassembled and washed with 0.1%
Cats were sedated by intramuscular injection of ketamine- bovine serum albumin (BSA) (Sigma) in PBS. Cells were
HCl (Fort Dodge Laboratories, Fort Dodge, Iowa), 15 fixed by sequential immersion in methanol and acetone at
mg/kg, and acepromazine (The Butler Company, Rochester, -20°C for 10 min, washed in 0.1% BSA, dried, and stored at
N.Y.), 0.15 mg/kg. Peritoneal exudate cells were then ob- -70°C to await indirect IFA labelling.
tained by lavage of the peritoneal cavity with 300 ml of Ab preparations found to enhance FIPV UCD1 infection
sterile phosphate-buffered saline (PBS), at room tempera- at any of the initial screening dilutions were further evalu-
ture, containing 200 ,ug of gentamicin sulfate (GIBCO Lab- ated for enhancement of both FIPV UCD1 and FIPV 79-
958 OLSEN ET AL. J. VIROL.

1146, using serial twofold dilutions from 1:32 to 1:1,048,576.

Because of the limited number of primary macrophages
available from a single peritoneal lavage procedure, evalua-
tion of the two strains of virus was usually conducted by
using macrophages collected on different days. Inoculation
of virus plus enhancing FIPV-specific feline antiserum
served as a positive control for enhancement and as a control
for day-to-day variation in the macrophages in all assays.
Indirect IFA labelling. Indirect IFA labelling of infected L. .
macrophages or CrFKC was conducted as previously de- 0
scribed (42), with slight modification. Briefly, slides were
incubated with a 1:2,000 dilution (in PBS) of FIPV-specific,
hyperimmune feline antiserum in a humidified chamber for 030
60 min at room temperature. The slides were then thor- E 1 A
oughly washed with 0.1% BSA in PBS and gently blotted
dry. Slides were subsequently incubated with a 1:50 dilution coI
(in PBS) of fluorescein isothiocyanate-conjugated goat anti-
feline IgG (Organon Teknika) for 30 min at 37°C, washed w 20~~~~~~
with 0.1% BSA, and counterstained with 0.002% Evans blue
stain.
Quantitation of infection enhancement. Following indirect cti~ ~ ~ niermdluin
IFA labelling, each slide was evaluated for the number of
infected (IFA-positive) cells per well. The degree of FIPV LU
~~~~~~~~~~~~~~~~U~~L
infection enhancement caused by each dilution of antiserum
or MAb (enhancement factor [EF]) was calculated as fol-
lows: the mean number of infected cells per well for the wells
infected in the presence of a given dilution of Ab was divided
by the mean number of infected cells per well for the wells
infected with virus alone (41).
Staphylococcal protein A blockade of the Fc portion of
enhancing Abs. Parallel enhancement assays were conducted
in the presence or absence of soluble staphylococcal protein
A (P 6650; Sigma) as previously described (41). Protein A
was added to preincubated virus-Ab mixtures (1:512 dilution primary feline peritoneal macrophages by FIPV-specific feline anti-
serum. Primary feline peritoneal macrophages were cultured on
of antiserum, 1:1,024 dilution of ascites fluid) to a final eight-chamber glass Lab-Tek slides (Nunc) and infected with FIPV
concentration of 200 ,ug/ml. Virus-Ab mixtures were made in 79-1146 or FIPV UCD1 in the presence or absence of serial dilutions
media without fetal bovine serum to avoid competition for of FIPV-specific feline antiserum. EF values were calculated as the
protein A binding. The protein A-containing mixtures were mean number of infected cells per well (n = 3) for wells infected in
incubated for an additional 60 min at 37°C before inoculation the presence of FIPV-specific Ab divided by the mean number of
of macrophages. infected cells per well (n = 3) for the wells infected with virus alone.
Construction of RPV-S. A recombinant raccoonpox virus
expressing the S protein of FIPV 79-1146 (RPV-S) was

Downloaded from https://journals.asm.org/journal/jvi on 02 January 2025 by 45.78.52.78.

constructed (25) essentially as described by Mackett et al. SPF cat 22 days after infection with FIPV UCD1 (Fig. 1).
(19, 20). Briefly, a full-length clone of the S gene of FIPV The peak level of enhancement of strain UCD1 occurred at
79-1146, kindly supplied by M. C. Horzinek (Utrecht, The a serum dilution of 1:2,048. At this dilution of antiserum,
Netherlands) as a BamHI fragment in pGS20 (46), was there was a 56-fold increase in the number of macrophages
inserted into the tk gene of raccoonpox virus. Recombinant infected compared with macrophages infected with virus
viruses were selected by plaque titration in TK- 143B cell alone. The peak EF for strain 79-1146 was 34 and occurred at
monolayers under overlay medium containing 30 ,ug of a serum dilution of 1:8,192. Each data point represents the
bromodeoxyuridine (Boehringer Mannheim, Indianapolis, mean EF calculated from triplicate wells. (The remainder of
Ind.) per ml. Presence of the FIPV S gene in the recombinant the enhancement data [Fig. 4 through 10] is presented in
virus was determined by dot blot hybridization (21). S similar graphic form, although the scales on the axes vary to
protein expression was determined by immunoblotting (29), accommodate the data.) Infection was not detected among
RIPA (40), and indirect IFA labelling. macrophages inoculated with medium alone or with antise-
CrFKC were grown to confluence on eight-chamber Lab- rum plus medium. In addition, no effect on the level of
Tek slides and infected with RPV-S at a multiplicity of infection was noted in the presence of preimmune (corona-
infection of 3. Twenty-four hours after infection (at the time virus antibody-negative) serum.
of optimal plaque formation), the slides were processed for To reduce the inherent variability involved in using pri-
indirect IFA labelling using each of the MAbs assayed for mary macrophages, all of the data contained in Fig. 1 were
enhancement. derived by using macrophages obtained from a single donor
cat. However, the ability of polyclonal feline antiserum to
RESULTS enhance FIPV infectivity was confirmed by using a variety
of antisera and using macrophages from several SPF cats
ADE of FIPV infectivity by virus-specific feline antiserum. (data not shown).
ADE of FIPV infection of primary feline peritoneal macro- Murine immunoglobulin and feline Fc receptor compatibil-
phages was demonstrated by using serum obtained from an ity. Before attempting to document ADE by using MAbs, it
VOL. 66, 1992 ANTIBODY-DEPENDENT ENHANCEMENT OF FIPV 959

a TABLE 1. MAbs tested for ADE FIPV infection of primary

feline peritoneal macrophages
FIPV Virus
MAb Immunizing protein neutralization ADEc
viruSa specificity titerb
2A10 FIPV 79-1146 S <10 -
3G7 FIPV 79-1146 S 2,560 +
1A4 FIPV 79-1146 S <10 -
lAll FIPV 79-1146 M <10 -
3H8 FIPV 79-1146 S <10 -
3C3 FIPV 79-1146 S <10 -
1.2 FIPV 79-1146 S <10 -
2.1 FIPV 79-1146 S <10 -
3C7 FIPV 79-1146 S <10 -
15A9.9 FIPV 79-1146 M <10 -
16C11.3 FIPV 79-1146 N <10 -
16F5.11 FIPV 79-1146 S <10 -
16F7.2 FIPV 79-1146 S <10 -
17B7.1 FIPV 79-1146 N <10 -
b 17D7.4 FIPV 79-1146 S 10 +
17E1.6 FIPV 79-1146 S 10 +
17H12.1 FIPV 79-1146 S <10 -
18A7.4 FIPV 79-1146 S 2,560 +
18A9.4 FIPV 79-1146 S 40 +
18E7.6 FIPV 79-1146 S <10 -
18F2.11 FIPV 79-1146 S 10 +
18G1.7 FIPV 79-1146 S <10 -
18H9.1 FIPV 79-1146 S <10 +
19A7.6 FIPV 79-1146 N <10 -
19G11.10 FIPV 79-1146 S 20 +
20A5.4 FIPV 79-1146 S <10 -
20C6.10 FIPV 79-1146 S 15 +
23A1.8 FIPV 79-1146 S 10 +
23B4.9 FIPV 79-1146 S <10 -
23C1.2 FIPV 79-1146 M <10 -
23F4.5 FIPV 79-1146 S 2,560 +
23F8.1 FIPV 79-1146 S 320 +
FIG. 2. SRBC phagocytosis assay. Primary feline peritoneal 24H5.4 FIPV 79-1146 S 80 -
macrophages, cultured on eight-chamber glass Lab-Tek slides, were 52D5d FIPV DF2 N <10 -
inoculated with a 1% suspension of murine anti-SRBC-coated G4. ld FIPV DF2 M <10 -
SRBCs (a) or uncoated, fresh SRBCs (b). The functional interaction G7. ld TGEV Miller N <10 -

of the Fc portion of murine immunoglobulin with the Fc receptors H11.ld TGEV Miller M <10 -

present on feline macrophages is demonstrated by the prominent a

Strain of virus used to immunize mice for MAb induction. TGEV,
erythrophagocytosis of the Ab-coated SRBC (a). transmissible gastroenteritis virus.
b Assayed on CrFKC and expressed as the reciprocal of the highest dilution

Downloaded from https://journals.asm.org/journal/jvi on 02 January 2025 by 45.78.52.78.

of MAb (as ascites fluid) that completely inhibited cytopathology. A value of
<10 indicates that there was no evidence of virus neutralization at the highest
concentration of MAb tested. c +, ability of a given MAb to induce ADE of
was necessary to determine whether murine immunoglob- FIPV infection (EF of >2); -, lack of such ability (EF of <2).
ulins are functionally compatible with the Fc receptors d Kindly supplied and characterized (immunoblotting in place of RIPA) by
present on feline macrophages. Such compatibility was S. Fiscus, Fermenta Animal Health.
demonstrated by using murine anti-SRBC serum in an Ab-
coated SRBC phagocytosis assay. Primary feline macro-
phages phagocytosed murine anti-SRBC-coated SRBC (Fig. infected CrFKC. Those MAbs which reacted with the pox-
2a) but did not engulf uncoated SRBC (Fig. 2b). virus plaques (Fig. 3) were designated as S protein specific,
ADE of FIPV infectivity by virus-specific murine MAbs. and those that did not were considered to be M protein
Each MAb was assayed by indirect IFA labelling of FIPV- specific. Membrane protein-specific MAbs provided by S.
infected CrFKC before evaluation for enhancement ability. Fiscus served as controls for this experiment.
Only those MAbs demonstrating strong IFA signals to both Only 12 of the 37 MAbs tested were found to be capable of
FIPV 79-1146 and UCD1 were selected for further evalua- mediating ADE of FIPV infectivity (ADE+; Table 1). All of
tion. The characteristics of the 37 MAbs used in this study the enhancing MAbs were S protein specific, and all but one
are shown in Table 1. (18H9.1) of the enhancing MAbs had a virus neutralizing
The MAbs produced during this study were initially as- titer of .10. Interestingly, MAb 18H9.1 was the least
sayed by RIPA to determine FIPV protein specificity. Many enhancing and the only MAb to demonstrate enhancement of
of the M- and S-specific MAbs were found to precipitate only one of the two strains of FIPV tested. While the peak
both the M and S glycoproteins. The addition of various EF of 18H9.1 was only 5, this degree of enhancement was
reducing agents to the reaction mixture did not enhance our consistent in a repeated experiment. Conversely, all but one
ability to distinguish specificity to these two proteins by (24H5.4) of the neutralizing MAbs were also found to be
RIPA. To confirm which MAbs were S protein specific, the enhancing. All of the MAbs defined as nonenhancing had an
MAbs were tested by indirect IFA labelling of RPV-S- EF of <2.0 (with the majority of them having an EF of <1.2)
960 OLSEN ET AL. J. VIROL.

0 30
LU2
FIG. 3. CrFKC infected with a recombinant racoonpox virus co0~~~~~~~s
(RPV-S) expressing the S protein of FIPV 79-1146. CrFKC were
grown to confluence on eight-chamber glass Lab-Tek slides, in-
fected with RPV-S (multiplicity of infection of 3), and fixed 24 h after
infection. The poxvirus plaque shown was labelled by indirect IFA Ci
using MAb 23F4.5 and is representative of results obtained with the
S protein-specific MAbs.
1 ~~~~~~~~~~~
_-u- .U--. _

at all dilutions tested. There correlation between

was no
G>O C,) L>O v- NfCLJ
c,)
>sv e _ (D

neutralization or enhancement and IgG subclass (data not Nj It

C~U)OOO~-C~-- U)c0
"t
If
shown).
Seven of the twelve enhancing MAbs were completely
characterized in enhancement assays over 16 serial dilutions Ascites fluid dilutions
of ascites fluid, using both FIPV 79-1146 and FIPV UCD1 FIG. 4. ADE of FIPV (strains 79-1146 and UCD1) infection of
(Fig. 4 through 10). As in Fig. 1, all of the data contained in primary feline peritoneal macrophages by MAb 18H9.1. Methods
Fig. 4 through 10 were derived by using macrophages from and EF calculations are as described for Fig. 1.
the same donor cat in order to reduce macrophage variabil-
ity. To ensure that these MAbs were not uniquely enhancing
for macrophages from this one particular cat, each of the enhancing Ab, often to below even the level obtained with
enhancing MAbs was tested at a single dilution of 1:4,096 virus alone. This finding indicated that the enhancement of
(near peak EF for each) for its ability to enhance FIPV infection seen in the presence of Ab was Fc mediated.
UCD1 infection of macrophages which were collected and
processed the same day from two different SPF cats. When DISCUSSION
the results from these two cats were compared, the absolute

Downloaded from https://journals.asm.org/journal/jvi on 02 January 2025 by 45.78.52.78.

levels of enhancement for a given MAb were different, but Although the immunologic aspects of FIP in vivo have
the relative pattern of enhancement among the MAbs was been well documented since 1980 (13, 49), ADE of FIPV
consistent-MAb 18H9.1 was the least enhancing, 3G7 and infection of macrophages in vitro has been investigated only
17E1.6 were moderately enhancing, and 18A7.4, 23A1.8, recently (41). The experiments reported here were designed
23F4.5, and 23F8.1 were strongly enhancing. to address this in vitro phenomenon more completely and to
Infection in the presence of those MAbs showing the begin to define the specific viral components responsible for
highest EFs (18A7.4, 23A1.8, 23F4.5, and 23F8.1) also led to the induction of enhancing ABs.
the formation of dramatic syncytia among infected macro- ADE of virus infectivity in other virus systems has been
phages. Syncytia were most prevalent under conditions of quantified by using several different methods, including
near-maximal enhancement, suggesting that polykaryon for- determination of the number of cells infected by IFA label-
mation may simply have been due to the presence of a large ling (44) or infectious center assays (37) or by titration of the
number of infected macrophages in close proximity to one amount of virus (38), antigen (44), or reverse transcriptase
another. An epitope-specific basis for syncytium formation activity (10) generated. We chose to evaluate the number of
cannot be ruled out, however, since occasional syncytia macrophages infected during a single cycle of infection. In
were found in the presence of these MAbs at lower levels of doing so, we are evaluating primarily the enhancing effect of
enhancement, whereas no syncytia were formed among Ab at the level of virus uptake and initiation of infection. In
macrophages infected in the presence of the other enhancing contrast, if levels of total virus output from populations of
MAbs or in the absence of any Ab. cells infected in the presence or absence of Ab are com-
Enhancement is blocked in the presence of staphylococcal pared, the effect of Ab both at the cell surface and through-
protein A. Results of parallel infections carried out in the out virus replication must be considered. Previous reports of
presence or absence of staphylococcal protein A are shown ADE of virus infectivity have not provided a consensus as to
in Table 2. The presence of protein A did not adversely affect whether measurement of the number of cells infected and the
the infectivity of FIPV itself. However, protein A dramati- titer of virus produced lead to the same assessment of
cally reduced the level of FIPV infectivity in the presence of enhancement (12, 36).
VOL. 66, 1992 ANTIBODY-DEPENDENT ENHANCEMENT OF FIPV 961

a. 4U'
40 40
h_
0
0

E
~30 E 30-
U

wU 20 WU 20

10- 10 I,

CM
Ln * M~* 'P- --
0
) CMtU.
N~~0(LON00 - (O 00
cor-(

Ascites fluid dilutions c are as decie fo Fig. 1.

and cluaion Ascites fluid dilutions
FIG. 5. ADE of FIPV (strains 79-1146 and UCD1) infection of
primary feline peritoneal macrophages by MAb 3G7. Methods and FIG. 6. ADE of FIPV (strains 79-1146 and UCD1) infection of
EF calculations are as described for Fig. 1. primary feline peritoneal macrophages by MAb 17E1.6. Methods
and EF calculations are as described for Fig. 1.

We began by examining the ability of FIPV-specific feline

antiserum to mediate ADE of infection. Our experiments, 23F4.5 (Fig. 5 through 8, respectively). This phenomenon at
using the same assay but different virus stocks, different low dilutions of Ab has also been documented in prior
antisera, and different macrophage donor cats, confirm pre- studies of ADE of virus infectivity (6, 23, 41, 44). FIPV is
vious results (41) and demonstrate that ADE of FIPV typical of other virus systems in that Ab-mediated virus
infectivity is a real and reproducible phenomenon. The uptake and infection occur via an Fc receptor-mediated

Downloaded from https://journals.asm.org/journal/jvi on 02 January 2025 by 45.78.52.78.

results of our study of MAb-dependent enhancement of endocytosis pathway which can be blocked by lyso-
FIPV infectivity clearly confirm the hypothesis that en- motrophic amines (41). During endocytosis, there are both
hancement is mediated by Abs to specific sites on the S uptake and fusion phases of the internalization process. The
protein of FIPV. These data also provide the first in vitro bell-shaped enhancement profiles may reflect a need to
correlate to the results of a previous in vivo study in which optimize virion and Ab molecule concentrations for complex
vaccination of cats with a recombinant vaccinia virus ex- formation before uptake can occur. At high concentrations
pressing the S protein of FIPV 79-1146 caused ADE of of Ab, the virions may be so coated With Ab that they are
disease (45). MAbs have previously been used to define the unable to approach and fuse with the endosome membrane
viral proteins responsible for inducing enhancing Abs in due to steric hindrance (7, 8, 27), thus resulting in a level of
other virus systems (17, 24, 35). While it is generally infection below that seen with virus alone. Although this
assumed that ADE of virus infectivity is mediated by Ab to prozonelike effect at high concentrationsof Ab may reflect
surface proteins, it was important to test MAbs not only to virus neutralization, we have avoided that term since it is
the S and M proteins but also to the N protein of FIPV, since possible that the inhibition of virus infectivity by Ab in the
enhancement of dengue virus infectivity has been demon- endosome of a macrophage and virus neutralization in the
strated with a MAb specific for a nonenvelope protein (9). more traditional sense may occur by different mechanisms
The overall bell-shaped distribution of enhancement factor (7).
relative to dilution of antiserum or ascites fluid is consistent MAbs 23A1L8 (Fig. 9) and 23F8.1 (Fig. 10) typify the
with the results of other studies of ADE (6, 23, 26, 41). Our remaining seven enhancing MAbs in this study which did not
results also show that in some instances, low dilutions/high show a prozonelike effect to dilutions of 1:32. Enhancement
concentrations of Ab can actually reduce the level of infec- assays using these seven MAbs were extended to a dilution
tion to below that obtained with virus alone, resulting in an of 1:8. At this dilution, all of the MAbs except 23F8.1 still
EF of <1. This pattern is most easily seen in Fig. 4 (MAb induced substantial enhancement. (The level of infection in
18H9.1) but was also evident with polyclonal feline antise- the presence of a 1:8 dilution of MAb 23F8.1 was similar to
rum (Fig. 1) and with MAbs 3G7, 17E1.6, 18A7.4, and that seen with virus alone.) It is unclear whether these MAbs
962 OLSEN ET AL. J. VIROL.

oE~~~~~~~~~~~~~~~~~~~~~~~'IA i V \
60

C -D 40C

60- 20

C14CO 16D' W 4t CO~~ (9 W Dc

40
/~~~~0 0) 0 CO 11 o r
.C_
E.C
40_-
I ~~~~~~In
ID~~. ~~~~~~
._ . .
*-
_ _t
~ ~
co
t
~
*-*C-_
QeOOO-C
Ccs e Z D C\,l
)O
t) v-
_m co
CM a
co

Ascites fluid dilutions Ascites fluid dilutions

FIG. 8. ADE of FIPV (strains 79-1146 and UCD1) infection of
FIG. 7. ADE of FIPV (strains 79-1146 and UCD1) infection of primary feline peritoneal macrophages by MAb 23F4.5. Methods
primary feline peritoneal macrophages by MAb 18A7.4. Methods and EF calculations are as described for Fig. 1.
and EF calculations are as described for Fig. 1.

ing Abs and the binding avidity of each Ab for the two strains
are functionally different from the others, are specific for a of FIPV.
different epitope(s), or are different only by virtue of their Perhaps the most paradoxical of our results is the strong
concentration in the ascites fluid or their avidity for FIPV. It correlati'on between a MAb's ability to mediate both neutral-
is interesting, however, that enhancement by these MAbs ization and ADE of FIPV infectivity. It is important to
was also relatively less efficiently blocked by protein A e.mphasize that the neutralization ability of the MAbs in this

Downloaded from https://journals.asm.org/journal/jvi on 02 January 2025 by 45.78.52.78.

(MAbs 23A'1.8 and 23F8.1; Table 2), suggesting a function- study'was defined in a standard virus neutralization assay
ally different basis for their infection-enhancing effect. using CrFKC and not a's the prozonelike effect that occurred
Inspection of the enhancement profiles also reveals that at high conc'entrations of Ab in the macrophage assay
FIPV strains 79-1146 and UCD1 are enhanced to different system. As such, it is likely that the ability of Ab to mediate
degrees by most of the Abs evaluated. Strain variation has both neutralization and enhancement is dependent on the
previously been demonstrated for enhancement of dengue specific epit'ope involved and not simply the con'centration of
virus infection with MAbs (8, 22, 24) and for enhancement of Ab. Previous investigations of ADE have also shown that a
FIPV infection with antiserum (41). Morens et al. have given Ab can both neutralize and enhance virus infectivity.
debated whether such strain variation in ADE may reflect a The most exhaustive data come from work done with dengue
basic difference between each strain's ability to infect mac- virus (8, 23). Halstead et al. examined neutralization by
rophages in the absence of Ab (23, 24). Although strain plaque reduction neutralization in LLC-MK2- cells and en-
variation in the ability of FIPV 79-1146 and UCD1 to infect hancement in the murine macrophage cell line P388D1, using
macrophages has previously been documented (43), our seven different strains of dengue 2 virus and five MAbs (8).
results over repeated experiments did not demonstrate a While certain MAb and strain combinations demonstrated
significant difference in the ability of FIPV UCD1 and FIPV only neutralization or enhancement (similar to MAbs 18H9. 1
79-1146 to infect macrophages in the absence of Ab (data not and 24H5.4; Table 1), most MAbs could both neutralize and
shown). In addition, the pattern of strain variation of ADE enha'nce virus infectivity (8).
was not consistent among all of the enhancing MAbs. Four This description of ADE of FIPV infectivity, like that of
MAbs enhanced UCD1 more than 79-1146 (MAbs 18H9.1, ADE in other virus systems, does not resolve the question of
3G7, 17E1.6, and 18A7.4; Fig. 4 through 7, respectively), how Ab binding to a particular site on the virion'induces an
while two MAbs enhanced 79-1146 more than UCD1 (MAbs overall increased efficiency of infection. Among other theo-
23F4.5 and 23F8.1; Fig. 8 and 10, respectively). It is there- ries, antibody binding may facilitate uncoating (15), protect
fore likely that the strain variation in enhancement seen here against pH- or protease-initiated damage during endolyso-
is related to the specific epitopes recognized by the enhanc- somal transit, or enhance conformational changes required
VOL. 66, 1992 ANTIBODY-DEPENDENT ENHANCEMENT OF FIPV 963

180 i
-i UCD1
160 -
--iU-- 79-1146

140

O 120-
0
0

q~80- .se100-

E E
60- 80
co 60
c
I 60 A
LU~~~~~~~~~

rn
0 nl....................
. . ..-
Z
CM
Ct~CM ~ ~~I
t
cq
a) CD
Um co t
'0O0N t
r@
g
CO CO CM
co r-
CM
co
co (O cm 04~~~~' ODJ
___-
o, ,-.iOo,-X C.~~~~~ _ _ .

Ascites fluid dilutions Ascites fluid dilutions

FIG. 9. ADE of FIPV (strains 79-1146 and UCD1) infection of FIG. 10. ADE of FIPV (strains 79-1146 and UCD1) infection of
primary feline peritoneal macrophages by MAb 23A1.8. Methods primary feline peritoneal macrophages by MAb 23F8.1. Methods
and EF calculations are as described for Fig. 1. and EF calculations are as described for Fig. 1.

for virus fusion to the endosomal membrane. FIPV is unlike and cleavage of these S proteins by exogenously supplied
many of the other coronaviruses because its S protein is not trypsin has been shown to enhance FIPV infection of pri-
cleaved during synthesis in routine cell culture (1, 18). mary feline macrophages (1). Since the S protein is impor-
However, a trypsin-sensitive cleavage site has been bio- tant for viral attachment and membrane fusion of coronavi-
chemically identified for FIPV 79-1146 and FECV 1683 (1), ruses, it is possible that antibody binding to a critical site on

Downloaded from https://journals.asm.org/journal/jvi on 02 January 2025 by 45.78.52.78.

the S protein may facilitate cleavage or stabilize a subse-
quent conformational change required for uptake of FIPV.
TABLE 2. Effects of preincubation of virus-Ab complexes with The results of this work indicate that ADE of FIPV
staphylococcal protein A on ADE of FIPV infectivity infection of primary feline macrophages can be demon-
strated by using both virus-specific feline antiserum and
Mean no. of macrophages murine MAbs. The characteristics of ADE of FIPV infectiv-
Inoculum infected/Lab-Tek well ity outlined here parallel those of previously defined systems
- Protein Al + Protein A" such as dengue virus. In view of these results and the
well-documented occurrence of ADE of FIP in vivo, FIPV
FIPV UCD1 56 67 may serve as a useful model for future investigations into the
FIPV UCD1 plus: mechanism of ADE of virus infectivity and the correlation
Feline antiserumC 872 15
between enhanced infection in vitro and enhanced disease in
3G7d 507 20
17E1.6d 825 3 vivo.
18A7.4d 4,197 46
18H9.ld 90 12
23A1.8d 4,110 1,546 ACKNOWLEDGMENTS
23F4.5d 247 24
23F8.ld 6,833 1,479 This work was supported by a resident research grant from Solvay
Animal Health, Inc., and by private donations to the Cornell Feline
a
Macrophages were infected with FIPV UCD1, with or without Ab, in the Health Center. A portion of the MAbs listed in Table 1 was
absence of protein A. produced with the support of a grant from Rhone Merieux.
Macrophages were infected with FIPV UCD1, with or without Ab, after We gratefully acknowledge the technical support of C. Geissen-
preincubation with protein A at a final concentration of 200 p.g/ml for 60 min
at 37°C. ger, R. Adams, and M. Brush and are grateful to Jeffrey Barlough
c FIPV-specific feline antiserum dilution = 1:512. and Margaret Barr for reviewing the manuscript. We also thank
d Ascites fluid dilution =
1:1,024. Cheryl Stoddart for many helpful discussions.
964 OLSEN ET AL. J. VIROL.

REFERENCES 22. Morens, D. M., and S. B. Halstead. 1987. Disease severity-

1. Baines, J. D. 1988. Ph.D. thesis. Cornell University, Ithaca, related antigenic differences in dengue 2 strains detected by
N.Y. dengue 4 monoclonal antibodies. J. Med. Virol. 22:169-174.
2. Barlough, J. E., and C. A. Stoddart. 1990. Feline coronaviral 23. Morens, D. M., and S. B. Halstead. 1990. Measurement of
infections, p. 300-312. In C. E. Greene (ed.), Infectious diseases antibody-dependent infection enhancement of four dengue virus
of the dog and cat. W. B. Saunders Co., Philadelphia. serotypes by monoclonal and polyclonal antibodies. J. Gen.
3. Corapi, W. V., R. 0. Donis, and E. J. Dubovi. 1990. Character- Virol. 71:2909-2914.
ization of a panel of monoclonal antibodies and their use in the 24. Morens, D. M., C. N. Venkateshan, and S. B. Halstead. 1987.
study of the antigenic diversity of bovine viral diarrhea virus. Dengue 4 virus monoclonal antibodies identify epitopes that
Am. J. Vet. Res. 51:1388-1394. mediate immune infection enhancement of dengue 2 viruses. J.
4. Fiscus, S. A., and Y. A. Teramoto. 1987. Antigenic comparison Gen. Virol. 68:91-98.
of feline coronavirus isolates: evidence for markedly different 25. Ngichabe, C. K., C. W. Olsen, and F. W. Scott. Unpublished
peplomer glycoproteins. J. Virol. 61:2607-2613. data.
5. Gimenez, H. B., H. M. Keir, and P. Cash. 1989. In vitro 26. Ochiai, H., M. Kurokawa, K. Hayashi, and S. Niwayama. 1988.
enhancement of respiratory syncytial virus infection of U937 Antibody-mediated growth of influenza A NWS virus in mac-
cells by human sera. J. Gen. Virol. 70:89-96. rophagelike cell line P388D1. J. Virol. 62:20-26.
6. Gollins, S. W., and J. S. Porterfield. 1984. Flavivirus infection 27. Ochiai, H., M. Kurokawa, Y. Kuroki, and S. Niwayama. 1990.
enhancement in macrophages: radioactive and biological studies Infection enhancement of influenza A Hi subtype viruses in
on the effect of antibody on viral fate. J. Gen. Virol. 65:1261- macrophage-like P388D1 cells by cross-reactive antibodies. J.
1272. Med. Virol. 30:258-265.
7. Gollins, S. W., and J. S. Porterfield. 1986. A new mechanism for 28. Olsen, C. W., and F. W. Scott. 1991. Feline infectious peritonitis
the neutralization of enveloped viruses by antiviral antibody. vaccination-past and present. Feline Health Top. 6:1-4, 8.
Nature (London) 321:244-246. 29. Panicali, D., S. W. Davis, R. L. Weinberg, and E. Paoletti. 1983.
8. Halstead, S. B., C. N. Venkateshan, M. K. Gentry, and L. K. Construction and characterization of live virus vaccines by
Larsen. 1984. Heterogeneity of infection enhancement of den- using genetically engineered poxviruses: biological activity of
gue 2 strains by monoclonal antibodies. J. Immunol. 132:1529- recombinant vaccinia virus expressing influenza virus haemag-
1532. glutinin. Proc. Natl. Acad. Sci. USA 80:5364-5368.
9. Henchal, E. A., J. M. McCown, D. S. Burke, M. C. Sequin, and 30. Pedersen, N. C. 1983. Feline infectious peritonitis and feline
W. E. Brandt. 1985. Epitopic analysis of antigenic determinants enteric coronavirus infections. Part 2. Feline infectious perito-
on the surface of dengue-2 virions using monoclonal antibodies. nitis. Feline Pract. 13:5-20.
Am. J. Trop. Med. Hyg. 34:162-169. 31. Pedersen, N. C., and J. F. Boyle. 1980. Immunologic phenomena
10. Homsy, J., M. Meyer, and J. A. Levy. 1990. Serum enhancement in the effusive form of feline infectious peritonitis. Am. J. Vet.
of human immunodeficiency virus (HIV) infection correlates Res. 41:868-876.
with disease in HIV-infected individuals. J. Virol. 64:1437-1440. 32. Peiris, J. S. M., S. Gordon, J. C. Unkeless, and J. S. Porterfield.
11. Homsy, J., M. Meyer, M. Tateno, S. Clarkson, and J. A. Levy. 1981. Monoclonal anti-Fc receptor IgG blocks antibody en-
1989. The Fc and not CD4 receptor mediates antibody enhance- hancement of viral replication in macrophages. Nature (Lon-
ment of HIV infection in human cells. Science 244:1357-1360. don) 289:189-191.
12. Hotta, H., A. S. Wiharta, and S. Hotta. 1984. Antibody-medi- 33. Porterfield, J. S. 1986. Antibody-dependent enhancement of
ated enhancement of dengue virus infection in mouse macro- viral infectivity. Adv. Virus Res. 31:335-355.
phage cell lines, Mkl and Mml (41802). Proc. Soc. Exp. Biol. 34. Reed, L. J., and H. Muench. 1938. Simple method of estimating
Med. 175:320-327. 50 per cent endpoints. Am. J. Hyg. 27:493-497.
13. Jacobse-Geels, H. E., M. R. Daha, and M. C. Horzinek. 1980. 35. Robinson, W. E., T. Kawamura, M. K. Gorny, D. Lake, J.-Y.
Isolation and characterization of feline C3 and evidence for the Xu, Y. Matsumoto, T. Sugano, Y. Masuho, W. M. Mitchell, E.
immune complex pathogenesis of feline infectious peritonitis. J. Hersh, and S. Zolla-Pazner. 1990. Human monoclonal antibod-
Immunol. 125:1606-1610. ies to the human immunodeficiency virus type 1 (HIV-1) trans-
14. Jacobse-Geels, H. E., M. R. Daha, and M. C. Horzinek. 1982. membrane glycoprotein gp4l enhance HIV-1 infection in vitro.

Downloaded from https://journals.asm.org/journal/jvi on 02 January 2025 by 45.78.52.78.

Antibody, immune complexes, and complement activity fluctu- Proc. Natl. Acad. Sci. USA 87:3185-3189.
ations in kittens with experimentally induced feline infectious 36. Schlesinger, J. J., and M. W. Brandriss. 1981. Antibody-medi-
peritonitis. Am. J. Vet. Res. 43:666-670. ated infection of macrophages and macrophage-like cell lines
15. Jolly, P. E., D. L. Huso, D. Sheffer, and 0. Narayan. 1989. with 17D-yellow fever virus. J. Med. Virol. 8:103-117.
Modulation of lentivirus replication by antibodies: Fc portion of 37. Schlesinger, J. J., and M. W. Brandriss. 1981. Growth of 17D
immunoglobulin molecule is essential for enhancement of bind- yellow fever virus in a macrophage-like cell line, U937: role of
ing, internalization, and neutralization of visna virus in macro- Fc and viral receptors in antibody-mediated infection. J. Immu-
phages. J. Virol. 63:1811-1813. nol. 127:659-665.
16. King, A. A., J. J. Sands, and J. S. Porterfield. 1984. Antibody- 38. Schlesinger, J. J., and M. W. Brandriss. 1983. 17D Yellow fever
mediated enhancement of rabies virus infection in a mouse virus infection of P388D1 cells mediated by monoclonal anti-
macrophage cell line (P388D1). J. Gen. Virol. 65:1091-1093. bodies: properties of the macrophage Fc receptor. J. Gen. Virol.
17. Krilov, L. R., L. J. Anderson, L. Marcoux, V. R. Bonagura, and 64:1255-1262.
J. F. Wedgewood. 1989. Antibody-mediated enhancement of 39. Scott, F. W. 1986. Feline infectious peritonitis and other feline
respiratory syncytial virus infection in two monocyte/macro- coronaviruses, p. 1059-1062. In R. W. Kirk (ed.), Current
phage cell lines. J. Infect. Dis. 160:777-782. veterinary therapy IX. W. B. Saunders Co., Philadelphia.
18. Lai, M. M. C. 1990. Coronavirus: organization, replication and 40. Smith, G. L., M. Mackett, and B. Moss. 1983. Construction and
expression of genome. Annu. Rev. Microbiol. 44:303-333. characterization of an infectious vaccinia virus recombinant that
19. Mackett, M., G. L. Smith, and B. Moss. 1982. Vaccinia virus: a expresses the influenza virus haemagglutinin gene and induces
selectable eukaryotic cloning and expression vector. Proc. Natl. resistance to influenza virus infection in hamsters. Proc. Natl.
Acad. Sci. USA 79:7415-7419. Acad. Sci. USA 80:7155-7159.
20. Mackett, M., G. L. Smith, and B. Moss. 1984. General method 41. Stoddart, C. A. 1989. Ph.D. thesis. Cornell University, Ithaca,
for production and selection of infectious vaccinia virus recom- N.Y.
binants expressing foreign proteins. J. Virol. 49:857-864. 42. Stoddart, C. A., and F. W. Scott. 1988. Isolation and identifica-
21. Mackett, M., G. L. Smith, and B. Moss. 1985. The construction tion of feline peritoneal macrophages for in vitro studies of
and characterization of vaccinia virus recombinants expressing coronavirus-macrophage interactions. J. Leukocyte Biol. 44:
foreign genes, p. 191-211. In D. M. Glover (ed.), DNA cloning, 319-328.
vol. II. IRL Press, Oxford. 43. Stoddart, C. A., and F. W. Scott. 1989. Intrinsic resistance of
VOL. 66, 1992 ANTIBODY-DEPENDENT ENHANCEMENT OF FIPV 965

feline macrophages to coronavirus infection correlates with in coronavirus spike proteins: implications for virus assembly. J.
vivo virulence. !. Virol. 63:436-440. Virol. 64:339-346.
44. Takeda, A., C. U. Tuazon, and F. A. Ennis. 1988. Antibody- 47. Weiss, R. C., and F. W. Scott. 1981. Antibody-mediated en-
enhanced infection of HIV-1 via Fc receptor-mediated entry. hancement of disease in feline infectious peritonitis: compari-
Science 242:580-583. sons with dengue hemorrhagic fever. Comp. Immunol. Micro-
45. Vennema, H., R. J. deGroot, D. A. Harbour, M. Dalderup, T. biol. Infect. Dis. 4:175-189.
Gruffydd-Jones, M. C. Horzinek, and W. J. M. Spaan. 1990. 48. Weiss, R. C., and F. W. Scott. 1981. Pathogenesis of feline
Early death after feline infectious peritonitis challenge due to infectious peritonitis: nature and development of viremia. Am.
recombinant vaccinia virus immunization. J. Virol. 64:1407- J. Vet. Res. 42:382-390.
1409. 49. Weiss, R. C., and F. W. Scott. 1981. Pathogenesis of feline
46. Vennema, H. L., L. Heijnen, A. Ziderveld, M. C. Horzinek, and infectious peritonitis: pathologic changes and immunofluores-
W. J. M. Spaan. 1990. Intracellular transport of recombinant cence. Am. J. Vet. Res. 42:2036-2048.

Downloaded from https://journals.asm.org/journal/jvi on 02 January 2025 by 45.78.52.78.