【病毒外文文献】1984 Experimental studies of a coronavirus and coronavirus-like agent in a barrier-maintained feline breeding colony

Archives of Virology 79 85 94 1984 Archives of Virology by Springer Vertag t984 Experimental Studies of a Corona irus and Coronavirus Like Agent in a Barrier Maintained Feline Breedinfl Colony By C A STOI DA I T J E BARLOUGtI and F W SCOTT Cornell Feline Health Center and Department of Microbiology New York State College of Veterinary Medicine Cornel1 University Ithaca New York U S A With 1 Figure Accepted August 15 1983 Summary MinimM disease MD cats raised in a barrier maintained feline breeding colony routinely become coronavirus CV antibody positive at 5 8 weeks of age and remain seropositive indefinitely In addition they shed corona virus like particles CVLPs in their feces which are morphologically distinct from typical coronaviruses 0ronasal intraperitoneal intraduodenal and intravenous inoculations of CVLPs were uniformly unsuccessful in the seroconversion of eats to CV whereas subsequent prolonged close contact with colony eats did result in seroconversion Such evidence indicates that MD eats in this barrier colony are infected with at least two non cross reactive agents 1 feline enteric CVLPs which are shed in feces and are morphologically and antigenicMly distinct from established coronaviruses CVs and 2 a virus which is antigenically cross reactive with CVs but which is apparently shed by some route other than the lower gastrointestinal tract Introduction Feline infectious peritonitis virus FIPV is a recognized member of the family Coronaviridae 2 18 and is antigenically related to transmissible gastroenteritis virus TGEV of swine canine coronavirus CCV and human respiratory coronavirus 229E HCV 229E 8 15 17 20 FIPV like particles have been identified in the gastrointestinal tracts of some cats with feline infectious peritonitis FIP 6 7 and enteritis 5 14 and in feces of apparently healthy cats N C PED RS personal communication An enteric virus morphologically and antigenically similar to FIPV has been reported from cats reared in a conventional feline breeding colony maintained 86 C A STODDART J E BARI OUGI I and F W ScoT by Animal Resources Services University of California Davis 14 and is referred to here as FECV UCD This virus was initially detected in feces of healthy cats but was shown to be capable of producing mild pyrexia leukopenia and enteritis when fed experimentally to young kittens 14 Coronavirus like particles morphologically distinct from FIPV have been detected in feces of clinically normal cats 10 11 and in feces of some cats with enteric disease 10 These latter particles were first visualized by electron microscopy EM in feces of MD cats raised in a barrier feline breeding colony maintained by the Division of Laboratory Animals Services New York State College of Veterinary Medicine Cornell University 11 Feline enteric CVLPs have since been detected by EM in 20 of 185 10 8 per cent fecal samples from privately owned cats hospitalized at t he small animal clinic of the Veterinary Medical Teaching Hospital at Cornell t0 While morphologic and antigenic similarities exist between FIPV and FECV UCD the morphologically dissimilar CVLPs appear to be antigenicMly dissimilar as well I 1 In this paper we describe transmission experiments involving feline enteric CVLPs and present evidence that MD cats in the Cornell University feline breeding colony are infected with at least two non cross reactive agents 1 feline enteric CVLPs which are shed in feces and are morphologically and antigenicMly distinct from the FIPV TGEV group and 2 a CV which is antigenicMly cross reactive with the FIPV TGEV group and which is apparently shed by some route other than the lower gastrointestinal tract Materials and Methods V ruse8 Feline enteric CVLPs were initially obtained from fresh fecal material collected from the MD feline breeding colony maintained by the Division of Laboratory Animal Services Cornell University An additional source was an experimental cat inoculated with CVLlos and which subsequently shed particles during the course of this study Fecal material was diluted 1 2 in 0 1 phosphate buffered saline PBS p 7 4 and clarified by centrifugation at 2000 g for I0 minutes at 4 C Supernatants were further clarified by ultraeentrifugation at 10 000 g for 10 minutes at 4 C and then pelleted at 110 000 g for i hour at 4 C Pellets were resuspended in 5 10 ml of either double distilled water DDW pH 7 0 for oronasal o n and intraduodenal i d inoculations or 1 BS for intraperitoneal i p and intravenous Lv inoculations This procedure produced highly concentrated C rLlo suspensions To sterilize the i p and i v inocula the loBS resuspensions were filtered successively through 800 and 450 nm filter units Sybron Nalge t oehester New York pelleted and suspended again as described above They were then filtered through 450 nm Aerodise filters Gehnan Sciences Inc Ann Arbor Iichigan Crude estimates of the number of particles mI of each inoeulum were made by EM 12 Inoeula were stored at 4 C for no longer than 24 hours prior to inoculation into experimental animals MD kittens from the Cornell feline breeding colony were used as a source of TGEV cross reactive CV for direct eat to eat transmission Coronavirus and Coronavirus Like Agent in Feline Colony 87 For morphologic comparison of CVs by EM the following additional viruses were used FECV UCD and FIPV WSU 79 1146 both kindly provided by N C Pedersen University of California Davis and TGEV Miller strain obtained from the New York State Diagnostic Laboratory Cornell University FECV UCD was supplied as a crude fecal suspension from infected eats while both FIPV and TGEV were supplied as clarified cell culture preparations Experimental Animals Nine 12 week old CV antibody negative and enteric CVLP negative MD kittens were purchased from a commercial breeding colony Liberty Laboratories Liberty Corner New Jersey Liberty eats Four 6 week old CV antibody negative MD kittens were obtained from the barrier maintained Cornell feline breeding colony CornelI cats Cats from the Corne l colony routinely become CV antibody positive at 5 8 weeks of age C A STODDA T unpublished data and remain seropositive indefinitely All animals were housed singly in negative pressure fiberglass isolation cages Germfree Laboratories Inc Miami Florida which were specially equipped for maximal air exchange Strict isolation procedures were followed in the care of eats throughout these experiments To further minimize cross contamination between groups experi mental and control eats were eared for on alternate days Electron Microscopy Fecal samples for electron microscopy were collected from litter pans and stored at either 4 or 80 C until preparation at which time a 30 50 percent suspension was made of each sample in DDW Samples were elarified by centrifugation at 2000 g for 10 minutes at 4 C further clarified by ultracentrifugation at 10 000 for 10 minutes at 4 C and then pelleted at 110 000 g for t hour at 40 C Pellets were resuspended in 5 10 drops of DDVg and stored at either 4 or 80 C For EM examination samples were negatively stained by mixing 1 drop of fecal suspension with 7 10 drops of 0 01 percent baeitracin and 2 4 drops of 3 5 percent phosphotungstic acid PTA pI I 7 0 Each mixture was then sprayed onto a carbon and parlodion coated 200 mesh copper grid with a glass nebulizer and examined in a Philips 201 electron microscope Individual samples were examined for an average of 12 minutes and no less than 6 minutes unless CVLPs were visualized sooner Samples were not scored positive unless 2 particles were found Serology Serum coronaviral antibody titers were determined using the Miller strain of TGEV in a kinetics based ELISA KELA as previously described 1 This assay has been shown to produce results which are comparable to the conventional heterologous immunofluorescence assay IFA 1 KELA titers 1 8 were considered negative Inoculation o Liberty Cats with Feline Enteric CVLPs Seven Liberty cats A G were inoculated with feline enteric CVLP suspensions by various routes Control cats H and I were given an equal volume of sterile diluent used to resuspend pellets Chronology of inoculations is shown in Table 1 Inoculation 6 was actually performed by natural exposure to an infected earner eat for 72 days For i d inoculation cats were anesthetized and abdominal incisions made in order to inject inoeulum directly into the duodenum 4 em posterior to the pylorie sphincter control inoculations were not performed for this experiment In inoculation 8 two concurrent doses were given one i v and one o n For i v inoculation CVLP sus pensions were injected into either the jugular or cephalic vein Fecal and serum samples were obtained from all eats before and after each inoculation Fecal samples were 88 C A STOI DAI J E BARLOIJG and F W SCOTT Table 1 Chronology o experimental inoculations o Liberty cats with dine enteric coronavirus like particles C VLPs Experi Inocu mental Control No of Source lation cats cats particles of no used used administered CVLPs 1 O O No of days doses oute observed interval of after between inoeu inocu doses lation a lation I A B C I I I 107 Colony 2 A C II I I06 Cat B 3 A I 107 Cat B 4 C I I 106 Cat B 5 D E G I I I i06 Cat B 6 b F ND c Undetermined Cat B 7 D E ND 106 Cat B 1 8 F G H I t07 Colony i 106 1 9 A B C I I 10 v Colony 1 D E 1 o n 50 4 o n 33 12 hours 5 o n 35 12 hours 1 i p 35 2 o n 35 24 hours Natural 60 exposure i d 48 i v 45 o n i v 21 a o n oronasal i p intraperitoneal i d intraduodenM i v intravenous b Natural exposure to an infected carrier cat eat B for 27 days ND not done collected and examined for CVLPs by EM 3 times weekly for the experimental cats and once weekly for control eats Serum samples were collected from each eat at approximate 2 week intervals Transmission oj Cornell Colony C V to Liberty Cats Each of 4 Cornelt MD kittens 1 4 was placed in the same cage with Liberty cats B D E and G respectively Two cats I t and I were used as room controls Serum samples were collected from all eats near or at the time of introduction and at 2 4 week intervals thereafter Kittens were separated from the Liberty eats after 55 days at which time individual fecal samples from all eats were collected and examined by EM Results Inoculation o Liberty Cats with Fdine Enteric CVLP8 Prior to inoculation all Liberty eat serum and fecal samples were CV antibody negative and CVLP negative respectively After inoculation 1 Table 1 cat B began shedding CVLPs in feces 14 days after inoculation Coronavirus and Coronavirus Like Agent in Feline Colony 89 This cat continued to shed particles intermittently throughout the course of this experiment and until the time of manuscript preparation a total period of 12 months Many CVLPs 50 i00 were observed within the first 1 2 minutes of EM search in most fecal samples which were CVLP positive CVLPs were visualized in only 2 fecal samples collected 47 and 49 days after inoculation from cat F after natural exposure to infected carrier cat B All fecal samples from control cats H and I were CVLP negative Throughout the inoculations all experimental and control cats remained CV antibody negative and no clinical signs of disease were seen Transmission o Cornell Colony C V to Liberty Cats Results from this experiment are shown in Table 2 Serum samples from all cats were CV antibody negative prior to introduction By 15 days after introduction 3 out of 4 75 percent of the Cornell colony cats had sero Table 2 Seroconversion o CV antibody negative Liberty cats by natural exposure to Cornell colony cats eciproeal CV antibody titer KELA b days after introduction CVLPs EM c days after introduction Cat a 11 0 15 45 71 i00 122 4 56 66 72 122 Cornel cats d 1 ND e neg net 2i 16 ND N D 2 ND neg 40 80 213 ND ND 3 ND neg 12 52 164 ND ND 4 ND neg 15 98 141 ND ND Liberty cats B neg ND 4 103 500 83 63 D neg ND 3 14 i0 23 33 E neg ND 5 30 104 65 238 G neg ND 4 43 51 78 97 Control cats H neg ND neg ND neg neg neg I neg ND neg ND neg neg neg ND ND ND ND ND ND ND ND ND ND ND ND Cat i pIaced with cat B 2 with D 3 with E 4 with G b KELA titers G 1 8 considered negative c Examination of feces for CVLPs by electron microscopy EM virus observed virus not observed a Cornell cats were 6 weeks old on the day of introduction and had not yet naturally seroeonverted Cornell colony cats routinely become CV antibody positive at 5 8 weeks of age and remain seropositive indefinitely e ND not done Cats H and I were housed individually in the same room under the same conditions as the experimental eats and served as room controls 90 C A STODDART J E BASLOUGIt a nd F W ScoTT converted while the Liberty cats remained seronegative By 45 days however all 4 Liberty cats had seroconverted as had the remaining Cornell cat no 1 There appeared to be little correlation between the magnitudes of CV antibody titers of cohabiting cats Fecal samples were not collected from the Cornelt cats prior to introduction but of the Liberty cats only cat B was shedding CVLPs prior to introduction After separation cats 1 and B 3 and E and 4 and G were shedding CVLPs Fecal samples from cats 2 and D were consistently CVLP negative Particles unequivocally recog nizable as coronaviruses of the same morphology as FECV UCD TGEV CCV and FIPV were never visualized in fecal samples Room control cats H and I remained CV antibody and CVLP negative and no clinical signs of disease were seen in any of the cats These results indicate that Cornell colony cats shed a second infectious agent in addition to enteric CVLPs which is cross reactive with TGEV and which appears to be transmitted by some route other than the lower gastro intestinaI tract Fig 1 Electron micrographs of negatively stained eoronavirus like particle and three typicM coronaviruses for morphologic comparison Bar represents 100 nm A feline enteric CVLP B FECV UCD C TGEV D FIPV Coronavirus and Coronavirus Like Agent in Feline Colony 91 Morphologic Comparison o CVLP and Typical CVs A comparison of CVLP and CV morphology is shown in Fig 1 The CVLP projections were markedly different from those of the three typical CVs FECV UCD FIPV and TGEV The peplomer fringe surrounding the CVLP was composed of regularly spaced teardrop like knobs anchored to the particle by thin stalks The total length of projections 25 nm was uniform The classical CV peplomers however were larger more diffuse and petal shaped The FECV UCD and TGEV peplomers were also not as regularly spaced and less uniform in length ranging from 15 30 nm Diseussion Enteric CVLPs are an enigmatic group of morphologically distinctive entities that have been identified by EM in feces of humans and a number of animal species including cats 11 13 Particles are moderately to markedly pleomorphie ranging in diameter from 60 to over 1000 nm and possess regularly spaced radiating surface projections reminiscent of coronaviral peplomers However closer examination reveMs that CVLP projections consist of a knoblike structure anchored to the particle by a slender stalk and thus are distinguishable from the larger more massive petal shaped peptomers of typieM eoronaviruses 4 13 Feline enteric CVLPs appear to be morphologically distinct from established eoronaviruses and also from a proposed feline enteric coronavirus FECV UCD recently reported from California 14 To date CVLPs have not been positively identified as the cause of disease in any species 13 Further immunologic and physieochemical studies are required before it can be determined whether CVLPs are true members of the Coronaviridae or represent a new and distinct group of viral agents In this paper we report that exposure of MD cats to feline enteric CVLPs did not result in the production of antibodies reactive with TGEV as determined in a kinetics based ELISA KELA Oronasal intraperitoneal intraduodenal and intravenous routes of administration were uniformly unsuccessful in seroconversion of eats This suggests that feline enteric CVLPs may be antigenieally distinct from the FIPV TGEV group of coronaviruses While remaining CV antibody negative after exposure to fecal material from colony eats MD cats nevertheless seroeonverted following prolonged close contact with colony kittens Most of these kittens while seronegative at the time of introduction became CV antibody positive within 15 days at which time their MD companion cats were still negative Table 2 All eats were seropositive by 45 days after introduction Moreover subsequent placement of one seroconverted MD eat together with one seronegative MD eat resulted in seroeonversion of the negative cat between 23 and 45 days 92 C A STODDART J E BARLOUGH and F W SCOTT after introduction data not shown The exact identity of the CV responsible for seroconversion has not been determined isolation of a CV from this colony has been unsuccessful and no true coronavirions have been observed by EM in an fecal specimens to date Data presented here suggest that fecal excretion of the TGEV eross reaetive CV as infectious virus if it occurs is minimal and that shedding probably occurs by some other route respiratory salivary etc After removal of colony kittens MD cats E and G began shedding CVLPs in theh feces Table 2 although they had not shed prior to introduction of the kittens even after repeated artificial inoculations Table 1 The kittens to which they had been exposed 3 and 4 also shed CVLPs suggesting that prolonged close contact with carrier eats may be required for efficient CVLP transmission Kitten 2 did not shed CVLPs after removal and its companion eat D also did not shed However both cats became CV anti body positive further supporting the conclusion that CVLPs and the sero converting CV are distinct There are additional possible explanations for failure of CVLPs to produce a TGEV cross reactive antibody response in MD cats For instance CVLPs may have been inactivated or antigenically altered during processing of inocula Viral peplomers presumed mediators of attachment to target cells may have been damaged or Iost producing loarticles incapable of infection EM examination of CVLPs however has demonstrated that peplomer integrity is virtuMty unaffected by the storage filtration ultra centrifugation and routine El processing teclmiques used in this experiment C A STODDA T unpublished data This is in contrast to some reports of established CVs in which such methods can result in damaged virions with extrusion of internal components and or partial or complete denudation of the peplomer fringe 9 C A STOD ART unpublished data The possibility also exists that subtle antigenic alteration of CVLP peptomers may not have been accompanied by detectable morphologic changes The action of proteo lyric enzymes in the gastrointestinal tract on the antigenic and morphologie stability of virus particles must also be considered t6 Detection of antibodies cross reactive with TGEV could also be dependent upon the serologic technique and or reagents employed Some reports indicate that porcine antiserum to TGEV cannot detect FIPV antigen suggesting one way antigenic cross reactivity between TGEV and FIPV 19 20 At this time it is unclear whether a similar one way serologic cross reaction might occur between TGEV and CVLPs and it is possible that while antibody to CVLPs may not detect TGEV in the KELA anti TGEV antibody may react with CVLPs in a reciprocal test Invasion of cells by CVLPs within the gastrointestinal tract may have produced only a low grade focal infection without a systemic antibody response Alternatively CVLP infection may have stimulated only IgA Coronavirus and Coronavirus Like Agent in Feline Colony 93 and or IgM production but this possibility is less likely due to the combined heavy chMn light chain specificity of the conjugate used in the KELA The antibody response induced by CVLP infection may have been so transient as to have been missed by the sampling intervals used However i v administration of CVLPs would presumably have