key: cord-1010573-gzcb6uqw
authors: Martinez, Mitzi L.; Weiss, Richard C.
title: Detection of feline infectious peritonitis virus infection in cell cultures and peripheral blood mononuclear leukocytes of experimentally infected cats using a biotinylated cDNA probe
date: 1993-03-31
journal: Veterinary Microbiology
DOI: 10.1016/0378-1135(93)90016-z
sha: 548fbde9442680310e833796e4cd9e9ed5a0feb7
doc_id: 1010573
cord_uid: gzcb6uqw

Abstract A dot blot hybridization assay, using a biotinylated cDNA probe, was able to detect feline infectious peritonitis virus (FIPV) RNA in Felis catus whole fetus (fcwf-4) cells infected with the FIPV isolates DF2, 79-1146, UCDI, and UCD2. The probe cross-hybridized in the dot blot assay with nucleic acid of a closely related feline coronavirus, feline enteric coronavirus (FECV)-79-1683. To construct the probe, a 2.5 kilobase cDNA, prepared from FIPV-DF2 genomic RNA, was molecularly cloned. The recombinant cDNA clone was digested with the restriction endonuclease Rsa I, and an 870 basepair Rsa I fragment was isolated from vector DNA by agarose electrophoresis and glassmilk purification. This fragment was complementary to the 3′ three fourths of the nucleocapsid gene. The hybridization probe was prepared by random primed labeling in the presence of biotin-11-dUTP. Using an avidin-alkaline phosphatase conjugate and chemiluminescent substrate detection system, virus could be detected in as few as 3000 infected cells. In an in vivo study, the probe was used to detect FIPV RNA in peripheral blood mononuclear leukocytes (PBML) isolated at various post-infection days (PID) from cats experimentally infected with the FIP-producing coronavirus isolate FIPV-79-1146 or FIPV-DF2. Viral RNA could be detected in as few as 12 000 PBML isolated from cats at PID 7 and in 50 000 PBML at PID 22. There was no consistent pattern, however, between hybridization results and prognosis or severity of disease at the time of sampling. Despite some cross-hybridization with FECV RNA, this probe should be useful for diagnosis of FIP, because cats infected with FECV most likely do not become viremic.

worldwide (Horzinek et al., 1977; Horzinek and Osterhous, 1979) . FIP in cats occurs in two clinically distinct forms, effusive or noneffusive. The effusive form is characterized by the accumulation of fluid in one or more of the body cavities, while the noneffusive form is characterized instead by development of granulomatous lesions in various organs (Pedersen, 1987b) . The noneffusive form of FIP can be difficult to diagnose clinically, because many feline diseases present in a similar manner (August, 1984) . The FIPV is related antigenically to and crossreacts serologically with feline enteric coronavirus (FECV), porcine transmissible gastroenteritis virus (TGEV), canine coronavirus (CCV), and the human coronavirus 229E ( Pedersen et al., 1978; Horzinek et al., 1982; Pedersen and Black, 1983) . Because cats become infected subclinically with various coronaviruses, serological crossreactions between FIPV and other coronaviruses complicate diagnosis of FIP using standard antibody tests (Barlough, 1984; 1985) . Homologous reactions after subclinical infections by FIPV may also complicate the interpretation of serological assays.

The FIPV genome consists of a positive-sense, single-stranded, infectious RNA of at least 27 kilobases (kb) (SiddeU et al., 1983; DeGroot et al., 1987b) . The 9.6 kb at the 3' end of the genome encode for three structural proteins: a 205 kilodalton (kDa) peplomer glycoprotein (S), a 45-50 kDa nucleocapsid protein (N), and a 25-32 kDa integral membrane protein (M) (DeGroot et al., 1987a; Tupper et al., 1987; Spaan et al., 1990) . The remainder of the genome is believed to encode RNA polymerase and other non-structural proteins .

Virulent FIPV isolates preferentially infect and replicate within monocytes and macrophages, while avirulent coronaviruses are considerably less able to infect mononuclear phagocytes (Pedersen, 1976; Stoddart and Scott, 1989 ). Weiss and Scott (198 la) demonstrated by animal inoculation studies that either whole blood or peripheral blood mononuclear leukocytes (PBML) from cats with FIP are infectious to other cats and can transmit FIP. Experimental infection of SPF cats with the non-FIP producing enteric virus, FECV, results in a limited distribution of virus from the primary site of infection, the mature columnar epithelial cells in the upper third of the villi in the small intestinal mucosa. Virus is demonstrated occasionally in monocytic cells within the medullary sinuses of the mesenteric lymph nodes (Pedersen et al., 1981; Pedersen, 1987; Evermann et al., 1988 ) ; in contrast to FIP, there is no evidence that FECV can induce a systemic disease (Pedersen et al., 1981 ) .

Apart from laparotomy and biopsy in noneffusive FIP, or clinical diagnosis of effusive FIP in the terminal stage, there are presently no specific antemortem diagnostic tests for FIP. A test which can detect FIPV directly in the blood or other clinical sample would be a valuable diagnostic tool. Direct detection of FIPV antigen in blood components with antibodies generally has not been successful, possibly because of low levels of viremia. Virus isolation is not useful for diagnosis because field isolates grow poorly in cell culture (Pedersen and Floyd, 1985 ) .

In this study, we describe a dot blot hybridization procedure, using a biotinylated recombinant cDNA probe complementary to a major portion of the N protein gene, to detect FIPV in feline cell cultures and PBML isolated from cats infected experimentally with the virulent FIP virus isolate 79-1146 or DF2.

Seven of the 9 cats used in these studies were healthy, 10-12 month-old specific-pathogen-free (SPF) males and females. The cats were purchased from a commercial breeder (Liberty Laboratories, Liberty Corners, N J) and were FeLV test-negative (by ELISA) and feline coronavirus antibody-negative prior to the studies. Two other cats (Z 10, Z 12 ) were 18 month-old males and were healthy and feline coronavirus antibody-positive prior to the studies. These cats originally were SPF kittens that had survived a previous experimental challenge exposure with FIPV. The cats were housed separately in cages located in an isolation facility of the Scott-Ritchey Research Center and were tested and cared for according to humane standards as set forth in the "Guide for the Care and Use of Laboratory Animals" (Publication No. 85-23, National Institutes of Health, Bethesda, MD). All experimental protocols were approved by a University animal welfare committee prior to the studies.

Feline coronaviruses used in this study included three virulent (i.e., FIPproducing) isolates (FIPV-DF2, FIPV-79-1146, FIPV-UCD1 ) and two avirulent strains (FIPV-UCD2, FECV-79-1683,) . The FIPV-79-1146, FIPV-UCD1, FIPV-UCD2 and FECV-79-1683 strains (McKeirnan et al., 1981; Pedersen and Floyd, 1985 ) were supplied by Dr. N.C. Pedersen, University of California, Davis. The FIPV-DF2 strain was obtained from the American Type Culture Collection, Rockville, Md.; No. VR-743. Felis catus whole fetus (fcwf-4) cells were supplied by Dr. N.C. Pedersen and were used to propagate all the viruses. The fcwf-4 cells were grown in Eagle's minimal essential media supplemented with 10% heat-inactivated fetal bovine serum, 100 U of penicillin/ml, 100/tg of streptomycin/ml, and 2.5 /tg of amphotericin/ml, along with 2 mM L-glutamine and 1% nonessential amino acids. Cultures were maintained in a humidified incubator at 37 °C in an atmosphere of air with 5% COz. 

The FIPV-DF2 polyadenylated RNA was isolated from virus-infected fcwf-4 cell pellets ( 108 cells) using the Fast Track System (Invitrogen, San Diego, CA) following the manufacturer's protocol. Four micrograms of FIPV-DF2 RNA were converted into RNA-cDNA hybrids by reverse transcriptase, using oligodeoxythymidine for priming. Resulting hybrids were converted into double-stranded DNA by RNase H together with DNA polymerase I and Escherichia coli DNA ligase and blunt-ended by T4 DNA polymerase. The double-stranded cDNA was ligated to BstX I linkers, inserted into the BstX I site of the pcDNAII phagemid (Invitrogen), and propagated in E. coli INV 1 alpha F'cells (Invitrogen). One recombinant cDNA clone, H 1A3, was characterized by restriction enzyme analysis and partial DNA sequencing. This clone represents 2500 bases near the 3' end of the FIPV-DF2 genomic RNA (Fig. 1 ). The recombinant phagemid containing H 1 A3 was digested with Rsa I and the fragments separated by agarose gel electrophoresis. An 870 bp fragment complementary to the 3' three fourths of the N protein gene ( Fig. 1 ) was excised and purified from the gel using glassmilk (Geneclean II, Bio 101, La Jolla, CA). This fragment was labeled with biotin by random primer extension using hexanucleotide primers and Klenow enzyme in the presence of biotin-1 I-dUTP (Tropix, Bedford, MA) to make the hybridization probe, FIPV-Rsa I.

Confluent monolayers of 3 day-old fcwf-4 cell cultures ( 150 cm 2 ) were infected with virus at a multiplicity of infection of 1. At 8 hours post-infection (PI), cells were harvested by trypsinization. The cells were pelleted, washed once with phosphate-buffered saline (PBS) and then re-pelleted. The cell pellets were frozen and stored at -80°C until use. An uninfected fcwf-4 culture was harvested as described and served as a negative control (uninfected fcwf-4 cellular RNA). Before use, the cells were thawed quickly, resuspended in PBS, and counted on a Neubauer hemocytometer. The cell concentration was adjusted to 10 6 cells/ml in PBS.

Whole blood was collected at various intervals by jugular venipuncture from SPF cats that had been injected intraperitoneally previously with either FIPV-79-1146 or FIPV-DF2 (Table 1 ) . Heparinized whole blood was diluted 1 : 1.67 

Preparation of cell dot blots was done as described by Paeratakul et al. ( 1988 ) . Two-fold serial dilutions of each cell suspension, ranging from either 106 to 3× l04 cells/ml for infected fcwf-4 cells or from 2× 106 to 1.5X 104 cells/ml for PBML, were made in PBS. A total of 100/tl of each cell dilution was applied to a 0.45 micron nylon membrane filter (Tropilon, Tropix), using a dot blot apparatus (BioDot, Bio Rad, Richmond, CA). The filters were fixed for 1 h at 4°C in 1% glutaraldehyde in a solution of 3% NaC1, 10 mM NaH2PO4, and 40 mM Na2HPO4 (pH 7.4). The filters were washed 3 times in proteolytic buffer (50 mM EDTA, 0.1 M Tris-HCL, pH 8.0) and were incubated for 30 min at 37 °C in 10 ml ofproteolytic buffer containing 20 #g/ ml proteinase K (Promega, Madison, WI ). After washing once in proteolytic buffer, the filters were air-dried and stored at room temperature until use.

The membrane filters were prehybridized for 1 h at 65 °C in hybridization buffer (0.5 % polyvinyl pyrrolidone, 1 mM EDTA, 1 M NaC1, 50 mM Tris, pH 7.5, 5% dextran sulphate and 2.0% sodium dodecyl sulphate) and were then hybridized for 18 h at 65 °C in fresh hybridization buffer containing 100 ng of heat-denatured FIPV-Rsa I probe. For chemiluminescent detection, posthybridization washing and visualization were performed according to the manufacturer's instructions, using a commercial kit (Southern-Light Chemiluminescent Detection System; Tropix, Bedford, MA). Following application of the chemiluminescent substrate, AMPPD, the filters were wrapped in plastic wrap and kept at room temperature for 1 h. Kodak XAR-5 X-ray film was exposed to the wrapped filter for 6 to 10 rain and then developed in an automatic processor. In some experiments, colorimetric detection was performed on the hybridized membrane after chemiluminescent detection. For colorimetric detection, post-hybridization washing and visualization were performed according to the manufacturer's instructions, using a commercial kit (BluGENE Nonradioactive Nucleic Acid Detection System; Bethesda Research Labs, Gaithersburg, MD). To minimize contamination by ribonucleases, all solutions were treated initially with diethyl pyrocarbonate (Sigma, St. Louis, Mo.) and all glassware was autoclaved and then baked at 250°C for at least 4 hr prior to use.

The 2.5 kb FIPV cDNA clone, H1A3, was oriented by restriction enzyme analysis (Fig. 1 ) and partial sequencing, using the sequence of the 3.2 kb at the 3' end of genomic RNA of FIPV-79-1146 (DeGroot et al., 1988 Vennema et al., 1990 ) as a reference. The position of the 870 bp Rsa I probe is indicated by the bold line in the lower diagram of Fig. 1 . Complementarity of the 870 bp Rsa I probe to the 2.5 kb FIPV cDNA insert was confirmed by Southern blotting (data not shown).

The FIPV-Rsa I probe hybridized to RNA in cells infected with each of the 5 feline coronaviruses tested (Fig. 2) . Viral RNA was detected in as few as 3000 cells for FIPV-DF2, FIPV-79-1146 and FECV-79-1683. Viral RNA was detected in 25 000 cells for FIPV-UCD2 and in 100 000 cells for FIPV-UCD 1. The probe did not hybridize above background levels to RNA isolated from normal uninfected feline cells.

The FIPV-Rsa I probe hybridized to RNA in PBML isolated from cats that had been infected with either FIPV-79-1146 or FIPV-DF2. Cats El, E2, and E3 were infected with FIPV-79-1146. The strongest hybridization signal occurred in cat E2 at PID 7 (Fig. 3 ) . The signal was detectable in as few as 6000 PBML. At PID 14, 25 000 PBML were required to produce a similar signal Table 1 .) PBML from an uninfected cat, Z6, are included as a negative control. Signal was detected using a colorimetric detection system as described in Materials and Methods. The PBML from cats infected with FIPV-DF2 were tested at PID 22 and 25. Two (DI, B6) of 3 cats tested at PID 22 showed a positive hybridization signal (Fig. 3) . At the time of testing, cat D1 had symptoms consistent with noneffusive FIP, and this cat died at PID 39. Cat B6, however, was only mildly anemic at the time of testing and presently is clinically normal (5 months post-infection). Cat D5, which tested negative at PID 22, had symptoms of noneffusive FIP and later died at PID 33 (Table 1 ) . A hybridization signal above the negative control was not detected in cat Z 10 or Z 12; both cats had been infected with FIPV-DF2 for at least a year prior to testing and were clinically healthy. A11 (6/6) of the FIPV-DF2-infected cats, including Zl0 and Z 12, had virus neutralizing (VN) antibody titers against FIPV-DF2. Four of these cats (DI, B6, D5, C1 ) were tested for serum FIPV antibody titers by indirect fluorescent antibody (IFA) assay, and all were positive (Table 2 ) .

The FIPV-Rsa I recombinant cDNA probe hybridized to fcwf-4 cells infected with each of several different feline coronavirus isolates. These isolates represented diverse pathotypes: the FIP-producing DF2, 79-1 146 and UCD 1, and the avirulent UCD2 and FECV-79-1683. As shown in Fig. 2 , the probe hybridized with equal intensity to both FIPV-DF2 and FIPV-79-1146. This finding was not unexpected, because the RNA sequence near the 3' end of the genome of these 2 isolates is very similar, if not identical (unpublished data). The decreasing intensity of signals seen for FECV-79-1683, FIPV-UCD 1 and FIPV-UCD2, compared to FIPV-DF2 or 79-1146, may have been due either to less homology between nucleotide sequences or possibly to a lower number of viral copies per cell, particularly with virus strains, such as UCDI, that replicate more slowly in cell culture (data not shown).

Experiments performed using less stringent hybridization conditions (hybridization and post-hybridization washes at 55°C) did not show significantly altered signal intensities. Due to functional constraints, there are other areas of the viral genome that may show greater sequence homology among isolates. Conceivably, the putative RNA polymerase gene at the 5'-end of the genome contains sequences that are widely conserved among coronaviruses. Comparison of amino acid sequences derived from the RNA polymerase gene of TGEV to amino acid sequences derived from the RNA polymerase gene of infectious bronchitis virus (IBV) revealed a high level of homology; this is in contrast to the low levels of homology between TGEV and IBV structural proteins (Britton, et al., 1990; Wesley, 1990) . Possibly, a probe constructed from the RNA polymerase gene can hybridize with more consistent intensity to the different coronavirus isolates tested.

Additionally, we wanted to determine if the probe would detect viral RNA in PBML isolated from cats experimentally infected with the FIP-producing isolates, FIPV-79-1146 or FIPV-DF2. As shown in Fig. 3 , the probe did hybridize to PBML isolated at various intervals from cats infected previously with either of the two isolates. Although the hybridization signal was stronger in cats infected with FIPV-79-1146, the samples were obtained at an earlier PID than were samples from cats infected with FIPV-DF2. All 3 cats infected with FIPV-79-1146 were positive at PID 7, which was during the incubation period for the experimental disease (Pedersen et al., 1984) . In general, the signal intensity decreased with time. In the FIPV-DF2-infected animals, results were variable at PID 22 and were negative in the single cat (C1) tested at PID 25. Overall, hybridization results were not correlated with prognosis or severity of disease at the time of sampling.

Cats Z10 and Z12 were included to determine if the probe would detect viral RNA in PBML from cats that had recovered from clinical FIP and possibly were latent carriers of the disease (Pedersen, 1987a) . A signal was not detected in PBML from these animals. If indeed they were latent carriers, then either the cDNA probe hybridization assay system was too insensitive, or the virus was not actually circulating in leukocytes but was instead sequestered within macrophages in lymph nodes or bone-marrow, as suggested previously by Pedersen and Floyd ( 1985 ) .

The results obtained with cats D5 and C 1 suggest that the dot blot hybridization format is not sensitive enough to detect the very low levels of viremia which may occur in this disease. Viremia in FIP is primarily cell-associated (Weiss and Scott, 198 la) . Conceivably, only very few circulating leukocytes may be infected, whereas macrophages may replicate FIPV very efficiently (Weiss and Scott, 1981 b) . To increase test sensitivity, we are currently evaluating two different methods to detect FIPV RNA in PBML: ( 1 ) a fluorescent in situ hybridization (FISH) of PBML cytospun onto glass slides and tested with biotinylated probes, and (2) reverse transcription of viral RNA isolated from PBML, followed by a polymerase chain reaction (PCR) technique, using oligonucleotide primers specific for FIPV to amplify the virus. Preliminary results obtained with FISH showed greater sensitivity than dot blot hybridization; PBML from D5 and CI, which were previously evaluated to be virus-negative, tested positive (data not shown).

In addition to studying FIP viremia, we are evaluating PBML isolated from cats infected with the non-FIP-producing coronaviruses, FECV-79-1683, FIPV-UCD2 and CCV. Previous studies indicate that cats infected with these viruses most likely have only a localized enteric infection and do not have a viremia (Pedersen et al., 1981; Barlough et al., 1984; Pedersen and Floyd, 1985 ) . PBML will be probed by FISH using our FIPV-cDNA probes to determine whether or not coronaviral RNA is present. If it can be shown that viral nucleic acids are in PBML of cats with FIP and are not in PBML from cats infected with non-FIP-producing coronaviruses, cDNA probes which crosshybridize between FIPV and FECV may still be valuable diagnostic tools.

Based on hybridization of a recombinant biotinylated FIPV-cDNA probe to PBML from FIPV-infected cats, an assay has been developed which can detect virus infection directly in blood samples. It is our opinion that this assay can be used to diagnose acute FIPV infections, particularly in viremic cats. Tests that can detect virus or virus-associated components (e.g., proteins or nucleic acids) are preferable to diagnostic tests based on serological reactions. This is particularly true in disease states such as FIP, where anticoronavirus antibody crossreactivity or homologous reactions after subclinical infections by FIPV complicate the interpretation of serological assays. Further refinements in sensitivity and specificity of the in situ hybridization assay should facilitate specific detection of FIPV RNA and result in a clinically useful diagnostic test for FIP.

Feline infectious peritonitis: an immune-mediated coronavirus vasculitis

Serodiagnostic aids and management practice for feline retrovirus and coronavirus infections

Cats, coronaviruses and coronavirus antibody tests

Experimental inoculation of cats with canine coronavirus and subsequent challenge with feline infectious peritonitis virus

Genomic organization of a virulent isolate of porcine transmissible gastroenteritis virus

cDNA cloning and sequence analysis of the gene encoding the peplomer protein of feline infectious peritonitis virus

lntracellular RNAs of the feline infectious peritonitis coronavirus strain 79-1146

Sequence analysis of the 3' end of the feline coronavirus FIPV 79-1146 genome: Comparison with the genome of porcine coronavirus TGEV reveals large insertion

Characterization of a feline infectious peritonitis virus isolate

Biological and pathological consequences of feline infectious peritonitis virus infection in the cheetah

Feline infectious peritonitis: A worldwide serosurvey

Feline infectious peritonitis virus

Antigenic relationships among homologous StlUClural polypeptides of porcine, feline and canine coronaviruses

Isolation of feline coronaviruses from two cats with diverse disease manifestations

A fast and sensitive method for detecting specific viral RNA in mammalian cells

Morphologic and physical characteristics of feline infectious peritonitis virus and its growth in autochthonous peritoneal cell cultures

Virologic and immunologic aspects of feline infectious peritonitis virus infection

Coronavirus diseases (coronavirus enteritis, feline infectious peritonitis)

Attempted immunization of cats against feline infectious peritonitis, using avirulent live virus or sublethal amounts of virulent virus

Experimental studies with three new strains of feline infectious peritonitis virus: FIPV-UCD2, FIPV-UCD3 and FIPV-UCD4

Antigenic relationship of the feline infectious peritonitis virus to coronaviruses of other species

An enteric coronavirus infection of cats and its relationship to feline infectious peritonitis

Pathogenicity studies of feline coronavirus isolates 79-1146 and 79-1683

The biology of coronaviruses

Immunoehemistry of viruses, 1I. The basis for serodiagnosis and vaccines

Intrinsic resistance of feline peritoneal macrophages to coronavirus infection correlates with in vivo virulence

Antigenic and biological diversity of feline coronaviruses: feline infectious peritonitis and feline enteritis virus

Primary structure of the membrane and nucleocapsid protein genes of feline infectious peritonitis virus and immunogenicity of recombinant vaccinia viruses in kittens

Pathogenesis of feline infectious peritonitis: Nature and development ofviremia

Pathogenesis of feline infectious peritonitis: Pathologic changes and immunofluorescence

Nucleotide sequence of the E2-peplomer protein gene and partial nucleotide sequence of the upstream polymerase gene of transmissible gaslroenteritis virus (Miller strain)