2008 Pathogenesis of feline enteric coronavirus infection

Journal of Feline Medicine and Surgery (2008) 10, 529e541doi:10.1016/j.jfms.2008.02.006

Pathogenesis of feline enteric coronavirus infection

Niels C Pedersen DVM, PhD1,2*, Claire E Allen BS

2, Leslie A Lyons PhD3

1Department of Medicine andEpidemiology, School of VeterinaryMedicine, University of California,Davis, CA 95694, USA2Center for Companion AnimalHealth, School of VeterinaryMedicine, University of California,Davis, CA 95694, USA3Department of Population Healthand Reproduction, School ofVeterinary Medicine, University ofCalifornia, Davis, CA 95694, USA

*Corresponding author. Present addressAnimal Health, Room 213, CCAH BldMedicine, University of California, Davþ1 - 5 3 0 - 7 5 2 - 7 4 0 2 , F a x : þ1 - [email protected]

1098-612X/08/060529+13 $34.00/0

Fifty-one specific pathogen-free (SPF) cats 10 weeks to 13 years of age wereinfected with a cat-to-cat fecaleoral passed strain of feline enteric coronavirus(FECV). Clinical signs ranged from unapparent to a mild and self-limitingdiarrhea. Twenty-nine of these cats were FECV na€ıve before infection andfollowed sequentially for fecal virus shedding and antibody responses overa period of 8e48 months. Fecal shedding, as determined by real-timepolymerase chain reaction (RT-PCR) from rectal swabs, appeared within a weekand was significantly higher in kittens than older cats. FECV shedding remainedat high levels for 2e10 months before eventually evolving into one of threeexcretion patterns. Eleven cats shed the virus persistently at varying levels overan observation period of 9e24 months. Eleven cats appeared to have periods ofvirus shedding interlaced with periods of non-shedding (intermittent orrecurrent shedders), and seven cats ceased shedding after 5e19 months (average12 months). There was no change in the patterns of virus shedding among catsthat were excreting FECV at the time of a secondary challenge exposure. Fourcats, which had ceased shedding, re-manifested a primary type infection whensecondarily infected. Cats with higher feline coronavirus (FCoV) antibody titerswere significantly more likely to shed virus, while cats with lower titers weresignificantly less likely to be shedding. Twenty-two kittens born toexperimentally infected project queens began shedding virus spontaneously, butnever before 9e10 weeks of age. Natural kittenhood infections appeared to below grade and abortive. However, a characteristic primary type infectionoccurred following experimental infection with FECV at 12e15 weeks of age.Pregnancy, parturition and lactation had no influence on fecal shedding byqueens. Methylprednisolone acetate treatment did not induce non-shedders toshed and shedders to increase shedding.

Date accepted: 29 February 2008 � 2008 ESFM and AAFP. Published by Elsevier Ltd. All rights reserved.

Feline enteric coronavirus (FECV) is a ubiq-uitous, worldwide, intestinal virus of cats(Pedersen et al 1981a, 2004). The name

feline coronavirus (FCoV) has been appliedsomewhat interchangeably to FECV. Technically,FCoV includes all strains (numerous), serotypes(types I and II) and biotypes (enteric or infectiousperitonitis viruses) of the genus. Several strainsof FECV have been studied by experimentalfecal-oral infection with cat-to-cat passed virus.The original FECV strain was designated FECV-University of California, Davis (UCD) (Pedersenet al 1981a) and a second isolate FECV-Rogers

: Center for Companiong., School of Veterinaryis, CA 95694, USA. Tel:5 2 - 7 7 0 1 . E - m a i l :

� 2008 ESFM a

and Morris (RM) (Hickman et al 1995). Both ofthese strains belong to serotype I, possessing a fe-line- rather than canine-coronavirus spike pro-tein. Several additional FECV strains have beenstudied in the field using polymerase chain reac-tion (PCR) (Foley et al 1997b, Benetka et al 2006).FECV is tropic for the mature apical epitheliumof the intestinal villi (Pedersen et al 1981a) andboth type I and II serotypes use species- andprobably type-specific (Dye et al 2007) variantsof aminopeptidase-N as a receptor (Tresnanet al 1996, Tusell et al 2007). FECV infection isusually unapparent or manifested by a transientgastroenteritis (Hayashi et al 1982, Pedersen et al1981a, Mochizuki et al 1999); it is rarely fatalwhen in its native biotype (Kipar et al 1998).

The importance of FECV as a primary intestinalpathogen is minimal. However, FECV commonly

nd AAFP. Published by Elsevier Ltd. All rights reserved.

mailto:[email protected]

530 NC Pedersen et al

mutates in vivo and at least one mutant form(ie, biotype) causes a highly fatal disease knownas feline infectious peritonitis (FIP) (Poland et al1996, Vennema et al 1998). The precise nature ofthe mutation that causes this change in virulencehas been variably ascribed to differences in thespike protein (Rottier et al 2005) or to non-synon-ymous or deletion mutations in the 3c (small en-velope) gene (Vennema et al 1998). The incidenceof the enteric / FIP biotype mutation followingFECV infection is unknown but may be as highas 20% (Poland et al 1996) and is more likely tomanifest clinically in kittens (Foley et al 1997a)or immunocompromised cats (Poland et al1996). FIP virus (FIPV) differs from its strictly in-testinal tropic FECV parent in its affinity for mac-rophages (Pedersen 1987, Stoddart and Scott1989). This altered tropism allows the virus to be-come a systemic pathogen of macrophages, andthe resultant disease involves a complex interac-tion between host cellular and humoral immu-nity and infected macrophages (Pedersen andBoyle 1980 Pedersen 1987).

Although there have been numerous studies ofFIPVs, studies of FECVs have been surprisinglyfew. Experimentation with FECV has been ham-pered by its lack of growth in tissue culture.Therefore, infection studies have often relied onextracts of feces from cats infected with cat-to-cat passed virus (Pedersen et al 1981a, Polandet al 1996). Although one report suggests thata cultured strain of FCoV, WSU-79-1683, is a pro-totypic FECV (Pedersen et al 1984b), this authornow believes it to be a tissue culture attenuatedrecombinant of canine and feline coronavirus.This is given support by the complex patternsof recombination that have been described forWSU-79-1146 (a highly virulent FIPV) andWSU-79-1683, which were both isolated fromthe same laboratory at the same time (Herre-wegh et al 1998). WSU-79-1683 also lacks the7b gene, which is intact in cat passed FECVs(Herrewegh et al 1995). Therefore, studies ofFECV should use biotype confirmed fecal pas-saged virus until a proper FECV is adapted totissue culture.

The present study was designed initially toprove that resistance and susceptibility to FECVinfection were under genetic control, just as ge-netics appears to play an important role inFIPV resistance (Foley et al 1997a). Young catswere infected with the RM strain of FECV andtheir patterns of fecal virus shedding quantifiedover extended periods of time by periodic sam-pling. Cats that stopped shedding the virus after

8e12 months were than bred to cats with a simi-lar profile, and cats that appeared to be long-term shedders were bred to chronic shedders.Their kittens were then infected with FECVat 10e23 weeks of age and the cycle continued.The goal was to create two bloodlines, one resis-tant and one susceptible. Once this was accom-plished, the genetic basis for resistance/susceptibility was to be determined. After morethan 3 years, it became apparent that FECV resis-tance and susceptibility may not be definable bysimple Mendelian genetics. Therefore, a decisionwas made to concentrate on what was learnedabout FECV pathogenesis.

Methods

Experimental animals

Twenty-nine FECV na€ıve cats, ranging from kit-tens to aged animals, were obtained from thespecific pathogen-free (SPF) breeding colony ofthe Feline Nutrition Laboratory, UCDavis. Catswere housed in the feline research facilities ofthe Center for Companion Animal Health(CCAH), UCDavis. Care was provided by staffof the CCAH under the supervision of the Centerfor Laboratory Animal Services, UCDavis. Stud-ies were done under United States Departmentof Agriculture required Institutional AnimalCare and Use Committee approved protocols.Males and females were not neutered for thisstudy. Select animals were chosen for breedingduring the course of the study and 22 kittensproduced from these mating’s added to thestudy over time.

Experimental infection

Cats were infected with 0.5 ml orally of a fecalextract (Poland et al 1996) of the RM strain ofFECV (Hickman et al 1995). The initial groupof cats was infected several days after acquisi-tion, while kittens reared during the studywere infected at 12e15 weeks of age and ob-served for signs of acute or chronic disease.Cats were housed in open rooms, with nomore than five animals per room. These groupsremained relatively stable, except when tomsor queens were transferred for breeding orqueens isolated for birthing and kitten rearing.Reasonable precautions were taken to limitspread of contaminated litter by caretakers; dis-posable coveralls, boots, foot baths, hand wash-ing, gloves were used.

531Feline enteric coronavirus infection

Quantitation of FECV shedding

FCoV RNAwas quantified using purification pro-cedures and specific primers reported by Gut et al(1999). Feces were collected by inserting standardcotton tipped swabs into the rectum prior to infec-tion and at 1 week intervals for at least 2 months,and then at 1e2 month intervals thereafter. RNAwas isolated from the swabs (van der Hoek et al1995). Five microliters of the purified RNA wasadded to 7 ml of PCR mixture containing 6 ml ofTaqMan One Step RT-Master Mix (Applied Bio-systems, Foster City, CA), 0.31 ml of MuLV/RNaseInhibitor, 0.24 ml each of forward and reverseprimers, and 0.10 ml of RNase-free water. The12-ml reaction went through a reverse transcrip-tase step for 30 min at 48�C and AmpliTaq Gold(Applied Biosystems, Foster City, CA) activationfor 10 min at 95�C. The samples were put through40 cycles of 95�C for 15 s and 60�C for 60 s for RNAamplification. PCR was performed using AppliedBiosystems (Foster City, CA) 7300 Real-time poly-merase chain reaction (RT-PCR) System and 7300System Software. The positive/negative cutoff ofthe assay was around 75e100 RNA transcripts/swab. Therefore, swabs that were negative at1� log 10 were considered negative. The numberof RNA transcripts per swab was considered equalto the number of viral particle (Gut et al 1999),given that each FECV particle contains only oneRNA transcript. There was no evidence for fecalinhibitors of the RT-PCR assay used in this study;SPF cat fecal samples were always negative, butbecame rapidly and progressively positive afterexperimental infection. Therefore, internal DNA(Monteiro et al 1997) or RNA (Escobar-Herreraet al 2006) fragment controls were not employed.

FCoV antibody tests

Serum antibody titers to FECV were undertakenwith an indirect fluorescent antibody (IFA)procedure (Pedersen 1976) using FIPV-UCD1infected Fcwf-4 cells (Pedersen et al 1981b). Cellswere grown in 12-well Teflon coated microscopicslides and infected with FIPV-UCD1 tissue culturefluid when three-quarters confluent. Slides wereharvested after 24e48 h and fixed in absolute ace-tone. Each serum was tested at 1:5, 1:25, 1:100,1:400 and 1:1600 dilutions in Hank’s buffered sa-line solution. Serum was allowed to react for 1 h,slides washed, and a 1:50 dilution of rabbit anti-cat IgG (Antibodies Incorporated, Davis, CA95616) was over layered for 1 h. Slides were thanwashed, stained with dilute Evan’s blue dye,and cover slips mounted with 1:1 glycerin:saline.

Slides were read on an indirect fluorescent micro-scope and the titer listed as the last dilution ofserum that still produced noticeable fluorescence.

Statistical analysis

Data was recorded on Excel spread sheets (Mi-crosoft Office 2003, Microsoft, Redmond, WA98004), and statistical analyses, when indicated,undertaken with JMP Statistical Discovery Soft-ware (SAS, Cary, NC 27513) (www.jmp.com/software/). Significance (P� 0.05) was deter-mined by the program’s Student t-test.

Results

Outcome of primary infection

Thirty-three cats were infected with FECVand fol-lowed sequentially for fecal virus shedding overa period of 14e48 months (Table 1). Twenty-ninecats were FECV na€ıve at the start of the study.Four of these cats (A01eA04) were born duringthe course of the study to project queens and,therefore, not FECV na€ıve, but were virus nega-tive at the time of primary infection. Fecal shed-ding rose within a week and remained atconsistently high levels of 1012e1016 particles/swab for 2e10 months (Figs 1e3; Table 1). Peak vi-rus levels tended to drop to levels of 106e109 par-ticles per swab in the secondary stage of infectionthat followed (Figs 1e3).

Three different patterns of virus shedding werenoted in the secondary infection stage. Eleven catsshed the virus continuously at greatly varyinglevels over an observation period of 14e24 months(persistent infection) (Table 1; Fig 1). Twelve catshad brief periods of recovery, interlaced with pe-riods of virus shedding (intermittent or recurrentshedders) (Table 1; Fig 2), and 10 cats ceased shed-ding at 7e18 months (average 12.3 months) (Table1; Fig 3). Three representative cats were graphedfor each of the three infection outcomes (Figs1e3). None of the cats developed FIP.

Outcome of secondary infection

Nineteen cats were used for this study and dividedinto two groups of four and 15 based on their virusshedding patterns prior to reinfection. The fourcats that had low or non-measurable virus shed-ding at the time of secondary exposure were clearlyreinfected. Fecal shedding for one of these cats is il-lustrated in Fig 4. Figure 5 shows the mean virus

Table 1. Description of 33 cats used to study patterns of fecal FECV shedding following primary infectionand in the study on the effect of methylprednisolone acetate induced stress in 18 of these animals

Catnumber

Gender/age(months)

Observationperiod (months)

Duration of primaryinfection (months)

Outcome ofinfection

Time to recovery(months)

94309 F/151 16 4 Persistent NA*94529 F/154 16 3 Recurrent NA98462 F/108 13 3 Recurrent NA99402 F/96 12 6 Recurrent NA00417 F/96 14 4 Recurrent NA01282 F/33 13 2 Recurrent NA02136y F/59 48 5 Recurrent NA04096 F/51 10 4 Persistent NA04139 F/52 13 10 Recurrent NA04140 F/52 13 8 Recurrent NA04141 F/52 10 2 Persistent NA04144 F/52 9 2 Persistent NA04146 F/52 9 3 Persistent NA04161 M/51 9 2 Persistent NA04224 M/50 9 4 Persistent NA04225 M/50 12 5 Persistent NA98272y M/90 24 4 Persistent NA04099z M/51 36 4 Recurrent NA05243y F/52 24 3 Recurrent NA05244z M/52 24 3 Recurrent NA05246y F/52 23 5 Recovered 1905249z F/52 23 5 Recovered 1805325z M/52 23 5 Persistent NA05326y M/52 23 5 Persistent NA06028y M/52 14 2 Recovered 1506029z F/53 14 2 Recovered 1506032y F/53 14 4 Recovered 1206033z F/53 14 2 Recovered 1206034y M/53 14 3 Recovered 15A01z M/2 13 4 Recovered 9A02y M/2 12 2 Recovered 11A03z F/2 16 3 Recurrent NAA04y F/2 8 5 Recovered 5

*NA¼ not applicable.yMethylprednisolone acetate treatment group.zNon-methylprednisolone acetate treatment group.


shedding levels for all four of the cats that were re-infectable; the peak levels of virus shedding wereas high as observed during primary infection andthe duration was similar (4e7 months). No evi-dence for reinfection was observed in cats thathad been shedding high levels of virus at thetime of secondary challenge exposure (Fig 6).

Relationship of age to peak virus sheddingduring primary infection

Cats were divided into three age groups: (1) kit-tens 2e4 months of age at the time of primary in-fection (n¼ 22), (2) mature cats 2e8 years of age(n¼ 25), and (3) aged cats 8e13 years of age

(n¼ 4). The peak level of virus shedding duringtheir primary phase of FECV infection was com-pared between groups (Fig 7). Kittens shed sig-nificantly higher peak levels of virus than cats2e8 years of age; virus shedding was also higherthan for aged cats, but this difference was notsignificant. Aged cats (8e13 years of age) alsotended to shed higher levels than 2e8 yearolds, but the difference was also not significant.

Relationship of serum antibody titersand virus shedding status

Antibodies to FCoV were measured sequentiallyby the IFA procedure in 16 animals over a period

1

100

10000

1E+06

1E+08

1E+10

1E+12

1E+14

0 2 2 3 4 5 6 7 8 9 10

0 2 2 3 4 5 6 7 8 9 10

Months following primary infection

FE

CV

p

articles p

er sw

ab

1

100

10000

1E+06

1E+08

1E+10

1E+12

1E+14


FE

CV

p

articles p

er sw

ab

1

100

10000

1E+06

1E+08

1E+10

1E+12

1E+14

1E+16

1E+18

0 1 2 3 3 5 6 7 8 9 10 12 13 14 15 16Months following primary infection

FE

CV

p

articles p

er sw

ab

Fig 1. Typical fecal FECV shedding patterns of cats demonstrating a persistent pattern of infection.


of 12e24 months. These cats were randomly se-lected from among the 33 animals whose infectioncourse had been established. A total of 241 timematched serum/feces samples were analyzed(Fig 8). FCoV antibody titers were significantly(P¼ 0.05) higher among cats that were virus shed-ders at the time of testing than in the group of catsthat were non-shedders. Conversely, cats withtiters of 1:25 and lower, as a group, were signifi-cantly (P¼ 0.05) more likely to be non-shedders.However, there was considerable overlap in titers

and virus shedding status among individual catsin the two groups; virus shedders and non-shedders were to be found in individuals withthe lowest (5e25) and highest (1600) titers.

Natural transmission to kittensborn to project queens

Twenty-two kittens were born to eight differentqueens, and data was available for 12 of themfor the first 24 weeks of their lives. None of

110

1001000

100001000001E+061E+071E+081E+091E+101E+111E+121E+131E+14

0 1 1 2 3 4 5 6 7 8 9 10 11 12 13 14Months following primary infection

FE

CV

p

articles p

er sw

ab

110

1001000

100001000001E+061E+071E+081E+091E+101E+111E+121E+131E+141E+15

0 1 2 2 3 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17Months following primary infection

FE

CV

p

articles p

er sw

ab

110

1001000

100001000001E+061E+071E+081E+091E+101E+111E+12

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 17 18 19 20 21 22 23 24 27 30 32 35 37 38Months following primary infection

FE

CV

p

articles p

er sw

ab

Fig 2. Typical fecal FECV shedding patterns of cats demonstrating an intermittant pattern of infection.


these 12 kittens shed FECV before 9 weeks ofage, while all kittens tested at 9e11 weeksof age were shedding as a result of natural ex-posure (Fig 9). However, the level of virusshedding was relatively low, from 103 to 108

particles/swab, and declined to very low ornon-detectable levels by 13e15 weeks (Fig 9).All of the kittens were infected with FECV-RM at 10e17 weeks of age (average 13 weeks)regardless of their prior FECV shedding status.The subsequent pattern of fecal virus sheddingresembled that observed during primary

FECV exposure in coronavirus na€ıve cats (Figs1e3, 9).

Effects of pregnancy, parturition andlactation on FECV shedding

Fecal virus shedding was measured for a period of12 weeks before and 12 weeks after parturition inseven queens and nine litters (Fig 10). There wasno significant difference in the levels of FECVshedding as a result of pregnancy, parturition orlactation.

110

1001000

100001000001E+061E+071E+081E+091E+101E+111E+121E+131E+14


FE

CV

articles p

er sw

ab

110

1001000

100001000001E+061E+071E+081E+091E+101E+111E+121E+13

0 12Months following primary infection

FE

CV

p

articles p

er sw

ab

110

1001000

100001000001E+061E+071E+081E+091E+101E+111E+121E+131E+14


FE

CV

p

articles p

er sw

ab

1 2 3 4 5 6 7 8 9 10 11 13 14 15 16

0 121 2 3 4 5 6 7 8 9 10 11 13 14 15 16

0 121 2 3 4 5 6 7 8 9 10 11 13 14 15 16

17

17

Fig 3. Typical fecal FECV shedding patterns of cats demonstrating a self-limiting (recovery) pattern of infection.


Effects of methylprednisolone acetatetreatment on fecal FECV shedding

Methylprednisolone acetate (5 mg/kg) was ad-ministered twice intramuscularly at a 3-week in-terval to 10 randomly selected cats from theproject; eight cohort cats were given saline(Table 1). There was no statistical change in the

levels of virus shedding post-treatment in catsgiven methylprednisolone acetate (Fig 11) or sa-line (data not shown). Furthermore, cats in eithergroup that were shedding at the time of treat-ment were not induced to shed more virus andcats that were non-shedders did not start shed-ding (data not shown).

1

100

10000

1E+06

1E+08

1E+10

1E+12

1E+14

1E+16

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24Months following primary infection

FE

CV

p

articles p

er sw

ab

Fig 4. Levels of virus shedding prior to and after reinfection (arrow).


DiscussionThe present study added to our understandingof the course of FECV infection in domesticcats. There was a distinct primary stage of infec-tion that lasted from 7 to 18 months; the highestlevel of virus shedding occurred during thisstage. This primary stage was resolved in oneof three manners: (1) recovery, (2) persistentshedding, and (3) recurrent or intermittent shed-ding. These findings corroborated earlier studiesof naturally occurring FECV infection. In a com-prehensive study of 275 purebred cats from sixcatteries, fecal samples were collected every1e3 months for a year and virus shedding quan-tified by RT-PCR (Foley et al 1997b). A large pro-portion of these cattery cats shed virus at anygiven time, but most manifested cycles of infec-tion and shedding. Similarly, Harpold et al(1999) found that all adult cats in an Abyssinian

110

1001000

10000100000

100000010000000

1000000001000000000

1e+101e+11

FE

CV

p

articles p

er sw

ab

-15 -12 -10 -4 -1 0 2 4 6 9 11Weeks post reinfection

Fig 5. One-way analysis of levels of fecal FECV sheddingin a group of four cats that were shedding very low ornon-detectable levels of virus prior to infection. Virus levelsfollowing reinfection were higher at all time points than theywere prior to infection, but because of the small group size,only weeks 3 and 4 were significantly different.

cattery shed virus in their feces at least once dur-ing the year and 4/15 cats were shedding greaterthan 75% of the time. Rohner (1999) reported thatFECV shedding dramatically decreased over 2years in a group of naturally infected cats. Herre-wegh et al (1997) studied the persistence andevolution of endemic FECV infection in a closedcat-breeding facility. Viral RNA was detected byRT-PCR in the feces and/or plasma of 36 of 42cats (86%) tested. Four of five infected catswere still shedding when tested 111 days later.Two cats were then placed in strict isolationand virus shedding was found to last up to 7months in one animal.

Persistent and recovered infections might beexplained by relative differences in the strengthof local gut immunity. However, an immunologicexplanation for the recurrent pattern of sheddingwas not so apparent. It would be tempting toblame periods of recurrent shedding on frequentreinfections. Reinfection is a common occurrence

1

100

10000

1000000

100000000

1e+10

1e+12

1e+14

1e+16

FE

CV

p

articles p

er sw

ab

-15 -12 -10 -4 -1 0 2 4 6 9 11Weeks post reinfection

Fig 6. One-way analysis of levels of fecal FECV sheddingin a group of 15 cats that were shedding virus at the timeof their secondary challenge exposure. There was no signif-icant change in virus shedding following reinfection.

Fig 7. One-way analysis of the mean peak levels of FECVfecal shedding during primary infection in cats infected at2e4 months of age, >2< 8 years of age, and >8 years of age.


for many gut pathogens, because local immunityoften requires persistence of the organism anddoes not possess strong memory (Brandtzaeg2007). However, successful reinfection in fourcats resembled a primary infection in magnitudeand duration, which was not true for recurrentbouts of shedding. It is possible that recurrentshedding was an artifact of the assay procedure.If the assay failed to delineate low level sheddingfrom non-shedding, recurrent and persistentshedders would be basically the same acceptfor amplitude. The alternative possibility wasthat these periods of reshedding were due to re-activation of a latent or sub-detectable infection.However, this was not supported by studies ofnatural or artificial stress (see below).

FECV was shed at very high levels followingprimary infection and the levels were signifi-cantly higher in kittens than in adult cats. Rohner(1999) also found that the levels of FECV were

Shedding status

010203040506070

5-25 100 400FECV antibody titer

1600

PositiveNegative

Nu

mb

er o

f sam

ples

Fig 8. FCoV indirect IFA antibody titers in serum collectedfrom cats over a 12e24 month period. Their fecal FECVshedding status was measured at the same time.

many log10 higher during early than late infec-tion. Foley et al (1997b) were the first to showhigher levels of FECV shedding in kittens thanolder cats in shelters. These findings have an im-portant collective implication for FIP. FIP is muchmore common among younger cats (Pedersen1976, Foley et al 1997a). If kittens are infected be-fore their immune systems are fully matured,levels of FECV replication would be higher, andgreater levels of virus replication would favorFECV / FIPV mutations. Relative age-relatedimmunodeficiency could also prevent kittensfrom containing the FIP mutant virus. This sce-nario is supported by research with FECV-RMinfection in cats that were immunocompromisedby long-standing FIV infection (Poland et al1996). Chronic FIV infected cats shed 10e100times more FECV than age-matched non-FIV in-fected cats, just like FECV in kittens, and 2/19 ofthem developed FIP vs none of the 20 FIV freecohorts. It was concluded that immunosuppres-sion caused by chronic FIV infection enhancedthe creation and selection of FIPV mutants by in-creasing the rate of FECV replication in thebowel, as well as by inhibiting the host’s abilityto combat the mutant viruses once they occurred.

The study also followed kittens born to FECVinfected queens. None of these kittens shed virusprior to 9 weeks of age, while all kittens thatwere tested between 10 and 15 weeks were pos-itive. These findings were in concordance withthose of Foley et al (1997a,b), who were notable to detect virus in feces before 10 weeks ofage in cattery kittens. However, Harpold et al(1999) found that kittens in an Abyssinian catterystarted shedding virus at 33e78 days (5e11weeks) of age (mean 9.6 weeks). Gut et al(1999) studied 77 kittens from 12 catteries andfound a progressive rise in fecal shedding fromaround 2e4 weeks onwards, with a peak at 9weeks of age. Therefore, the period of 9 weeksof age is probably when most kittens are in-fected, although it may occur at an earlier ageunder certain conditions. These chronologicalfindings support FCoV control programs that ad-vocate isolation of pregnant queens and earlyweaning of their kittens (Addie et al 2004). The-oretically, if queens are strictly isolated and theirkittens separated at the earliest possible time(<5e6 weeks of age), the kittens will remain,for the most part, free of FECV. The problemwith this procedure is the occasional infectionof kittens at 2e5 weeks of age (Gut 1999,Harpold et al 1999) as well as the problem of pre-venting infection after weaning. FECV is easily

0

2

4

6

8

10

12

14

16

8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24Weeks of age

FE

CV

sh

ed

din

g p

er sw

ab

(lo

g10)

C01C02D01E01E02E03

A01A02A03A04B01B02

Fig 9. Fecal virus shedding levels in kittens born to project queens. Kittens were infected naturally at 9e10 weeks of age,but this infection appeared transient. Kittens were experimentally infected at 10e17 weeks of age (average 13 weeks).


transmitted from room to room by caregiverseven with very good containment facilities andprocedures (Pedersen et al 1981a). The ease of fo-mite transmission and ubiquitous nature of thevirus makes it extremely difficult to keep catsfree of the virus. However, delaying FECV infec-tion until the kittens are older (>16 weeks), byisolation and early weaning, may have another

0

2

4

6

8

10

12

-12 -10 -8 -6 -4 -2 0Weeks after par

FE

CV

fecal sh

ed

din

g p

er sw

ab

(lo

g10)

Fig 10. Average levels of FECV fecal shedding before and a

depressing effect on FIP regardless of whetheror not they are infected later in life. The levelsof virus replication might be lower in older kit-tens, thus decreasing the likelihood of mutants,and the immune system would be more compe-tent in containing any FIPV that might arise.

Kittens born into the study, and infected natu-rally, demonstrated a peculiar form of infection.

2 4 6 8 10 12turition

5243

4140(1)

4140(2)

5246

5249(1)

5249(2)

4139

2136

6032

fter parturition in seven queens during nine pregnancies.

1

100

10000

1000000

100000000

1e+10

1e+12

1e+14

Viru

s p

artic

le

s p

er s

wa

b

-11 -9 -6 -1 0 1 2 3 4 5 6 7 8Weeks post injection of depoMedrol

Fig 11. Levels of fecal FECV in cats that were positive shed-ders and treated with two injections of methylprednisoloneacetate at week 0 and 4. There was no significant differencein levels of virus shed in the feces after treatment.


The levels of virus replication in naturally in-fected kittens were much lower than in olderkittens and cats that had been experimentallyinfected. The virus shedding also seemed to bemuch more transient. Unlike cats with longerterm infections, which did not respond to su-per-infection, kittens responded to an experi-mental challenge exposure just like na€ıve orrecovered cats. This suggests that maternal im-munity may have played some role in alteringthe course of natural infection, although notleading to either strong premonition or adaptiveimmunity.

The role of stress in reactivating possible latentor subclinical infections, or increasing virusshedding among shedders, was studied in twoways: (1) by studying a natural stress, ie, preg-nancy, parturition and lactation, and (2) inducingan artificial stress with corticosteroid treatment.The stress of parturition and lactation is knownto activate latent feline herpesvirus infection inabout 40% of queens, and this re-excretion of vi-rus is an important route for infection of kittens(Gaskell and Povey 1977). This stress can bemimicked by the administration of corticoste-roids (Gaskell and Povey 1977, Hickman et al1994); a single dose of methylprednisolone ace-tate is particularly effective (Reubel et al 1993).Methylprednisolone is also a potent activator oflatent feline leukemia virus infection (Rojkoet al 1982, Pedersen et al 1984a,b) and willabolish age-related resistance to FeLV (Pedersenand Johnson, 1991). It will even reactivatesubclinical dermatophyte infections in kittens inthe post-recovery phase (Pedersen 1991). Unlikefeline herpesvirus, neither parturition/lactationnor methylprednisolone treatment affectedFECV shedding in this laboratory setting.

It might be argued that conditions in nature aremore severe; however, FECV shedding was alsonot affected by parturition and lactation incattery cats (Foley et al 1997b).

There was a significant relationship betweenserum IFA titers and virus shedding in this study.Cats that were shedding virus tended to have IFAtiters of 1:100 and above. Cats that were no longershedding virus tended to have titers of 1:25 andlower. This confirmed an earlier report by Rohner(1999). However, such a relationship was notnoted by Harpold et al (1999) or insinuated byFoley et al (1997b). This discrepancy may involvethe manner in which data is viewed. When fecalshedders and non-shedders are looked at asgroups, the relationship between higher titersand shedding and lower titers and non-sheddingwas significant. However, the overlap between ti-ter and shedding was substantial and greatly di-minished the value of titers in evaluatingshedding status in individual cats. The value ofapplying group FECV.

Application of this technique to eradication ofFECV under experimental conditions was firstreported by Hickman et al (1995). A FECV wasinadvertently introduced into a very large re-search cat-breeding colony and not recognizeduntil a few cases of FIP were confirmed overthe next couple of years. In order to save valu-able blood lines, FECV was eradicated based onserum antibody titers. Cats with high or rising ti-ters were culled and cats with low and decreas-ing titers were taken out of breeding andmaintained in strict isolation. The process of con-tinually selecting cats with low titers eventuallyyielded a much smaller group of cats that wereproven to be free of FECV by following titers intheir kittens. If the titers in kittens were strictlyof maternal origin, they would become negativeby 12e16 weeks. If the kittens were infected,they would fall, and then rise again after 12e16weeks. Such an eradication regimen requires ex-ceptional quarantine facilities and infection con-trol practices and accurate titer determinations.These are difficult to achieve in most multi-catenvironments in the field. Although the use ofgroup titers for eradication may not be usefulto many multi-cat households and catteries,group titers may be helpful in looking at theoverall coronavirus status in cattery or othermulti-cat environments. Groups of cats thattend to have high titers are likely to have a signif-icant proportion of FECV shedders, while theconverse would be true for groups of cats thathave low and negative titers. Catteries with


many high titer cats are known to have a greaterincidence of FIP (Foley et al 1997a,b).

Immunity to FECV infection was not studiedper se, but it was possible to infer several thingsfrom the present observations. First, there isa definite primary stage of infection that lastsseveral months, followed by a period when virusshedding either remains persistent at a lowerlevel, becomes intermittent, or ceases. The immu-nity generated during this primary stage wasslow to develop, variable in strength, and tenu-ous in duration. Reinfection also appeared tomirror primary infection, indicating that immu-nity is not only tenuous, but that it lacks memory.The finding that immunity is tenuous and rein-fection common mirrors the conclusions ofAddie et al (2003), who found that FECV recov-ered cats can be reinfected with the same or dif-ferent strains of the virus. This is characteristic ofgut immunity in general (Brandtzaeg 2007) andto coronavirus immunity in particular (Saif2004). This pattern of infection and immunity isstrongly influenced by environmental factors.FECV infection would be self-limiting if groupsof cats were allowed to grow old without con-stant re-exposure to other infected cats and tonew cats (especially kittens). Certain husbandrypractices unique to cats may also favor cat-to-cat (ie, litterboxes and litter) and cat-to-human-to-cat (fomite) transmission.

AcknowledgementsThis study was funded by Winn Feline Founda-tion of the Cat Fanciers Association, Manasquan,NJ 08736-0805, and the Center for CompanionAnimal Health, School of Veterinary Medicine,University of California, Davis, CA 95616. Theauthors are also grateful for the technical supportgiven by Ms Kelly Bettencourt and Ms ElizabethHolmes.

ReferencesAddie DD, Schaap IA, Nicolson L, Jarrett O (2003) Persis-

tence and transmission of natural type I feline coronavirusinfection. Journal of General Virology 84 (10), 2735e2744.

Addie DD, Paltrinieri S, Pedersen NC, Second internationalfeline coronavirus/feline infectious peritonitis sympo-sium. (2004) Recommendations from workshops of thesecond international feline coronavirus/feline infectiousperitonitis symposium. Journal of Feline Medicine andSurgery 6 (2), 125e130.

Brandtzaeg P (2007) Induction of secretory immunity andmemory at mucosal surfaces. Vaccine 25 (30), 5467e5484.

Benetka V, Kolodziejek J, Walk K, Rennhofer M, Mostl K(2006) M gene analysis of atypical strains of feline andcanine coronavirus circulating in an Austrian animalshelter. Veterinary Record 159 (6), 170e174.

Dye C, Temperton N, Siddell SG (2007) Type I feline corona-virus spike glycoprotein fails to recognize aminopeptidaseN as a functional receptor on feline cell lines. Journal ofGeneral Virology 88 (6), 1753e1760.

Escobar-Herrera J, Cancio C, Guzman GI, Villegas-Sepul-veda N, Estrada-Garcıa T, Garcıa-Lozano H, Gomez-Santiago F, Gutierrez-Escolano AL (2006) Construction ofan internal RT-PCR standard control for the detection ofhuman caliciviruses in stool. Journal of Virological Methods137(2), 334e8.

Foley JE, Poland A, Carlson J, Pedersen NC (1997a) Riskfactors for feline infectious peritonitis among cats inmultiple-cat environments with endemic feline entericcoronavirus. Journal of the American Veterinary MedicalAssociation 210 (9), 1313e1318.

Foley JE, Poland A, Carlson J, Pedersen NC (1997b) Patternsof feline coronavirus infection and fecal shedding fromcats in multiple cat environments. Journal of the AmericanVeterinary Medical Association 210 (9), 1307e1312 (doctoralthesis, Univerity of Zurich).

Gaskell RM, Povey RC (1977) Experimental induction offeline viral rhinotracheitis virus re-excretion in FVR-recov-ered cats. Veterinary Record 100 (7), 128e133.

Gut M (1999) Kombination von Fruhabsetzen und Vakzinier-ung mit Primucell FIP. Evaluation der Wirksamkeit zurVerhinderung der Felinen Infektiosen Peritonitis.

Gut M, Leutenegger CM, Huder JB, Pedersen NC, Lutz H(1999) One-tube fluorogenic transcription-polymerasechain reaction for the quantitation of feline coronaviruses.Journal of Virological Methods 77 (1), 37e46.

Harpold LM, Legendre AM, Kennedy MA, Plummer PJ,Millsaps K, Rohrbach B (1999) Fecal shedding of felinecoronavirus in adult cats and kittens in an Abyssinian cat-tery. Journal of the American Veterinary Medical Association215 (7), 948e951.

Hayashi T, Watabe Y, Nakayama H, Fujiwara K (1982) Enter-itis due to feline infectious peritonitis virus. JapaneseJournal of Veterinary Science 44 (1), 97e106.

Herrewegh AA, Mahler M, Hedrich HJ, Haagmans BL, Eg-berink HF, Horzinek MC, Rottier PJ, deGroot RJ (1997)Persistence and evolution of feline coronavirus in a closedcat-breeding colony. Virology 234 (2), 349e363.

Herrewegh AA, Smeenk I, Horzinek MC, Rottier PJ, deGroot RJ (1998) Feline coronavirus type II strains 79-1683and 79e1146 originate from a double recombination be-tween feline coronavirus type I and canine coronavirus.Journal of Virology 72 (5), 4508e4514.

Herrewegh AA, Vennema H, Horzinek MC, Rottier PJ, deGroot RJ (1995) The molecular genetics of feline coronavi-ruses: comparative sequence analysis of the ORF7a/7btranscription units of different biotypes. Virology 212 (2),622e631.

Hickman MA, Reubel GH, Hoffmann DE, Morris JG,Rogers QR, Pedersen NC (1994) An epizootic of felineherpesvirus, type 1 in a large specific pathogen-free catcolony and attempts to eradicate the infection by identi-fication and culling of carriers. Laboratory Animals 28 (4),320e329.

Hickman MA, Morris JG, Rogers QR, Pedersen NC (1995)Elimination of feline coronavirus infection from a largeexperimental specific pathogen-free cat breeding colony


by serologic testing and isolation. Feline Practice 23 (3),96e102.

van der Hoek L, Boom R, Goudsmit J, Snijders F, Sol CJA(1995) Isolation of human immunodeficiency virus type1 (HIV-1) RNA from feces by a simple method and differ-ence between HIV-1 subpopulations in feces and serum.Journal of Clinical Microbiology 33 (3), 581e588.

Kipar A, Kremendahl J, Addie DD, Leukert W, Grank CK,Reinacher M (1998) Fatal enteritis associated with corona-virus infection in cats. Journal of Comparative Pathology 119(1), 1e14.

Mochizuki M, Osawa N, Ishida T (1999) Feline coronavirusparticipation in diarrhea of cats. Journal Veterinary MedicalScience 61 (9), 1071e1073.

Monteiro L, Bonnemaison D, Vekris A, Petry KG, Bonnet J,Vidal R, Cabrita J, Megraud F (1997) Polysaccharides asPCR inhibitors in feces: Helicobacter pylori model. Journalof Clinical Microbiology 35 (4), 995e998.

Pedersen NC (1976) Serologic studies of naturally occurringfeline infectious peritonitis. American Journal of VeterinaryResearch 37 (12), 1449e1453.

Pedersen NC, Boyle JF (1980) Immunologic phenomena inthe effusive form of feline infectious peritonitis. AmericanJournal Veterinary Research 41, 868e876.

Pedersen NC (1987) Virologic and immunologic aspects offeline infectious peritonitis virus infection. Advances Exper-imental Medicine and Biology 218, 529e550.

Pedersen NC (1991) Dermatomycosis (ringworm). In: FelineHusbandry, American Veterinary Publications Goleta, CApp. 254e261.

Pedersen NC, Johnson L (1991) Comparative efficacy ofthree commercial feline leukemia virus vaccines againstmethylprednisolone acetate-augmented oronasal chal-lenge exposure with virulent virus. Journal of the AmericanVeterinary Medical Association 99 (10), 1453e1455.

Pedersen NC, Boyle JF, Floyd K, Fudge A, Barker J (1981a)An enteric coronavirus infection of cats and its relation-ship to feline infectious peritonitis. American Journal ofVeterinary Research 42 (3), 368e377.

Pedersen NC, Boyle JF, Floyd K (1981b) Infection studies inkittens utilizing feline infectious peritonitis virus propa-gated in cell culture. American Journal of Veterinary Research42 (3), 363e367.

Pedersen NC, Meric SM, Ho E, Johnson L, Plucker S, TheilenGH (1984a) The clinical significance of latent feline leuke-mia virus infection. Feline Practice 14 (2), 32e48.

Pedersen NC, Evermann JF, McKeirnan AJ, Ott RL (1984b)Pathogenicity studies of feline coronavirus isolates 79-1146 and 79-1683. American Journal of Veterinary Research45 (12), 2580e2585.

Pedersen NC, Sato R, Foley JE, Poland AM (2004) Commonvirus infections in cats, before and after being placed inshelters, with emphasis on feline enteric coronavirus.Journal of Feline Medicine and Surgery 6 (2), 83e88.

Poland AM, Vennema H, Foley JE, Pedersen NC (1996) Tworelated strains of feline infectious peritonitis virus isolatedfrom immunocompromised cats infected with a felineenteric coronavirus. Journal of Clinical Microbiology 34(12), 3180e3184.

Reubel GH, Ramos RA, Hickman MA, Rimstad E, HoffmannDE, Pedersen NC (1993) Detection of active and latent fe-line herpesvirus 1 infections using the polymerase chainreaction. Archives of Virology 132 (3e4), 409e420.

Rohner M (1999) Bestimmung der Ausscheidungsfrequenzvon Felinen Cioronaviren unter Feldbedingungen,Doctoral thesis, University of Zurich.

Rojko JL, Hoover EA, Quackenbush SL, Olsen RG (1982) Re-activation of latent feline leukaemia virus infection. Nature298 (5872), 385e388.

Rottier PJ, Nakamura K, Schellen P, Volders H, Haijema BJ(2005) Acquisition of macrophage tropism during thepathogenesis of feline infectious peritonitis is determinedby mutations in the feline coronavirus spike protein.Journal of Virology 79 (22), 14122e14130.

Saif LJ (2004) Animal coronavirus vaccines: lessons forSARS. Developmental Biology (Basel) 119, 129e140.

Stoddart CA, Scott FW (1989) Intrinsic resistance of felineperitoneal macrophages to coronavirus infection corre-lates with in vivo virulence. Journal of Virology 63 (1),436e440.

Tresnan DB, Levis R, Holmes KV (1996) Feline aminopepti-dase N serves as a receptor for feline, canine, porcineand human coronaviruses in serogroup I. Journal of Virol-ogy 70 (12), 8669e8674.

Tusell SM, Schittone SA, Holmes KV (2007) Mutational anal-ysis of aminopeptidase N, a receptor for several group 1coronaviruses, identifies key determinants of viral hostrange. Journal of Virology 81 (3), 1261e1273.

Vennema H, Poland A, Foley J, Pedersen NC (1998) Felineinfectious peritonitis viruses arise by mutation fromendemic feline enteric coronaviruses. Virology 243 (1),150e157.

Available online at www.sciencedirect.com

2008 Pathogenesis of feline enteric coronavirus infection

Documents

Transcript of 2008 Pathogenesis of feline enteric coronavirus infection