CLINICAL MICROBIOLOGY REVIEWS, Jan. 1995, p. 87–112 Vol. 8, No. 1 0893-8512/95/$04.0010 Copyright q 1995, American Society for Microbiology Feline Immunodeficiency Virus: an Interesting Model for AIDS Studies and an Important Cat Pathogen² MAURO BENDINELLI, 1 * MAURO PISTELLO, 1 STEFANIA LOMBARDI, 1 ALESSANDRO POLI, 2 CARLO GARZELLI, 1 DONATELLA MATTEUCCI, 1 LUCA CECCHERINI-NELLI, 1 GINO MALVALDI, 1 AND FRANCO TOZZINI 2 Retrovirus Center and Virology Section, Department of Biomedicine, 1 and Department of Animal Pathology, 2 University of Pisa, Pisa, Italy INTRODUCTION .........................................................................................................................................................87 HIGHLIGHTS OF FIV EPIZOOLOGY ....................................................................................................................88 THE VIRUS AND ITS GENOME ..............................................................................................................................88 Morphology and Biochemistry of the Virion.........................................................................................................88 Genome Organization and Expression ..................................................................................................................90 Replication and Effects on Cells In Vitro..............................................................................................................93 THE INFECTION.........................................................................................................................................................94 Demonstration of FIV Infection..............................................................................................................................94 Host Invasion and Persistence ................................................................................................................................95 IMMUNE RESPONSES TO THE VIRUS ................................................................................................................96 Antigenic Determinants of FIV ...............................................................................................................................96 Antibody Responses ..................................................................................................................................................97 Cell-Mediated Responses .........................................................................................................................................98 PATHOLOGICAL EFFECTS ......................................................................................................................................98 Clinical Manifestations ............................................................................................................................................98 Acute phase ............................................................................................................................................................98 Asymptomatic phase .............................................................................................................................................98 Persistent generalized lymphadenopathy ..........................................................................................................99 ARC.........................................................................................................................................................................99 FAIDS .....................................................................................................................................................................99 Hematological, Immunological, and Other Laboratory Signs ............................................................................99 Histopathological Changes ....................................................................................................................................100 USEFULNESS OF FIV AS A MODEL FOR AIDS ...............................................................................................101 Getting a Better Understanding of Pathogenesis ...............................................................................................101 Developing and Improving Therapies ..................................................................................................................103 Criteria for Vaccine Development ........................................................................................................................103 CONCLUDING REMARKS ......................................................................................................................................106 ACKNOWLEDGMENTS ...........................................................................................................................................106 REFERENCES ............................................................................................................................................................107 INTRODUCTION When Niels Pedersen and coworkers in 1987 first described the feline immunodeficiency virus (FIV), they immediately realized its great potential as a model for human immunode- ficiency virus (HIV) studies. The virus was derived from a California domestic cat with clinical features that had led the owner to believe the pet had acquired some kind of AIDS, and it had the typical attributes of a lentivirus (174). The expecta- tion that FIV would become a leading animal model for HIV studies has since been fulfilled. Indeed, the similarities be- tween the two viruses are so many and profound that persons acquainted with HIV and HIV-induced disease will probably have a feeling of deja vu while reading this article. Subsequent investigations have shown that FIV is an impor- tant health problem for domestic cats throughout the world. Its existence had not been suspected before because, as in HIV infection, the underlying defect is a progressive disruption of the host’s immune functions and therefore the clinical mani- festations evolve slowly, are polymorphic, and are impossible to attribute to a single etiology in the absence of specific diagnostic tests. Given these premises, it is hoped that studying FIV will not only help us to understand the pathogenesis of HIV and provide clues for the development of interventive strategies for its control but also will benefit the natural host species. This review will principally address the life cycle of FIV in the cat, the immune responses evoked, and the clinical mani- festations associated with infection. The state of the art in current attempts to develop efficacious anti-FIV therapies and vaccines will also be a main focus, because it is thought that these attempts might teach a great deal about the best strate- gies to use in developing an anti-HIV armamentarium. Finally, the pros and cons of FIV as an animal model for AIDS studies will be examined in detail. Several more or less exhaustive reviews on FIV have already appeared (18, 23, 67, 77, 107, 123, 172, 173, 176, 214, 258). The present update is justified by the accelerating pace at which FIV research has been growing in recent years. * Corresponding author. Mailing address: Department of Biomedi- cine, Virology Section, Via San Zeno 37, I-56127 Pisa, Italy. Phone: 39 (50) 553562. Fax: 39 (50) 555477. ² This article is dedicated to the memory of Giovanni B. Rossi, who contributed immensely to the development of AIDS research in Italy and the European Community. 87 on February 11, 2020 by guest http://cmr.asm.org/ Downloaded from
Transcript
CLINICAL MICROBIOLOGY REVIEWS, Jan. 1995, p. 87–112 Vol. 8, No. 10893-8512/95/$04.0010Copyright q 1995, American Society for Microbiology
Feline Immunodeficiency Virus: an Interesting Model forAIDS Studies and an Important Cat Pathogen†
CARLO GARZELLI,1 DONATELLA MATTEUCCI,1 LUCA CECCHERINI-NELLI,1
GINO MALVALDI,1 AND FRANCO TOZZINI2
Retrovirus Center and Virology Section, Department of Biomedicine,1 and Department of Animal Pathology,2
University of Pisa, Pisa, Italy
INTRODUCTION .........................................................................................................................................................87HIGHLIGHTS OF FIV EPIZOOLOGY ....................................................................................................................88THE VIRUS AND ITS GENOME..............................................................................................................................88Morphology and Biochemistry of the Virion.........................................................................................................88Genome Organization and Expression ..................................................................................................................90Replication and Effects on Cells In Vitro..............................................................................................................93
THE INFECTION.........................................................................................................................................................94Demonstration of FIV Infection..............................................................................................................................94Host Invasion and Persistence................................................................................................................................95
IMMUNE RESPONSES TO THE VIRUS ................................................................................................................96Antigenic Determinants of FIV ...............................................................................................................................96Antibody Responses ..................................................................................................................................................97Cell-Mediated Responses.........................................................................................................................................98
Hematological, Immunological, and Other Laboratory Signs............................................................................99Histopathological Changes ....................................................................................................................................100
USEFULNESS OF FIV AS A MODEL FOR AIDS ...............................................................................................101Getting a Better Understanding of Pathogenesis...............................................................................................101Developing and Improving Therapies ..................................................................................................................103Criteria for Vaccine Development ........................................................................................................................103
When Niels Pedersen and coworkers in 1987 first describedthe feline immunodeficiency virus (FIV), they immediatelyrealized its great potential as a model for human immunode-ficiency virus (HIV) studies. The virus was derived from aCalifornia domestic cat with clinical features that had led theowner to believe the pet had acquired some kind of AIDS, andit had the typical attributes of a lentivirus (174). The expecta-tion that FIV would become a leading animal model for HIVstudies has since been fulfilled. Indeed, the similarities be-tween the two viruses are so many and profound that personsacquainted with HIV and HIV-induced disease will probablyhave a feeling of deja vu while reading this article.Subsequent investigations have shown that FIV is an impor-
tant health problem for domestic cats throughout the world. Itsexistence had not been suspected before because, as in HIVinfection, the underlying defect is a progressive disruption ofthe host’s immune functions and therefore the clinical mani-festations evolve slowly, are polymorphic, and are impossibleto attribute to a single etiology in the absence of specificdiagnostic tests. Given these premises, it is hoped that studyingFIV will not only help us to understand the pathogenesis of HIVand provide clues for the development of interventive strategiesfor its control but also will benefit the natural host species.This review will principally address the life cycle of FIV in
the cat, the immune responses evoked, and the clinical mani-festations associated with infection. The state of the art incurrent attempts to develop efficacious anti-FIV therapies andvaccines will also be a main focus, because it is thought thatthese attempts might teach a great deal about the best strate-gies to use in developing an anti-HIV armamentarium. Finally,the pros and cons of FIV as an animal model for AIDS studieswill be examined in detail. Several more or less exhaustive reviewson FIV have already appeared (18, 23, 67, 77, 107, 123, 172, 173,176, 214, 258). The present update is justified by the acceleratingpace at which FIV research has been growing in recent years.
* Corresponding author. Mailing address: Department of Biomedi-cine, Virology Section, Via San Zeno 37, I-56127 Pisa, Italy. Phone: 39(50) 553562. Fax: 39 (50) 555477.† This article is dedicated to the memory of Giovanni B. Rossi, who
contributed immensely to the development of AIDS research in Italyand the European Community.
Serologic and virus isolation studies have shown that FIV isenzootic worldwide. Its prevalence, however, varies greatlydepending on geographical location and other variables of thecat populations surveyed. Among apparently healthy domesticcats, the lowest seroprevalence rates (1% or less) were ob-served in central European countries and the United Statesand the highest (up to 30%) were seen in Japan and Australia.Age and gender also markedly affect FIV prevalence. Infectionis acquired most commonly after 1 year of age, and itsprevalence increases up to 10 years of age and then remainsstable or tends to decline (the mean lifespan of domestic catsis about 15 years), most likely as a result of a negative balancebetween FIV-induced mortality and new infections. Seroposi-tivity rates in male cats are two or more times higher than infemales. It is generally accepted that this and other variationsin FIV distribution are mainly the result of differences in socialbehavior and lifestyle of cats which influence the likelihood ofFIV transmission (see below). Another important variable thathas been seen to affect the results of serosurveys is the healthstatus of the animals. The seroprevalence rates observed wereusually several times higher in sick cats than in their healthycounterparts, thus providing a persuasive, though indirectindicator of the disease-inducing potential of FIV (9, 90, 98,123, 161, 172, 173, 176, 214, 255).Evidence of FIV infection has been obtained from cat sera
stored as far back as 1968 (202), and the virus is likely to haveaccompanied cats for a considerably longer time (212). Inter-estingly, although the means of FIV and HIV transmission areconsiderably different, the spread of FIV infection appears tohave a pattern similar to that observed for HIV infection inAfrica, where HIV probably had been prevalent long beforethe epidemic outbreak in western countries (200). It is notclear whether cats represent the only known reservoir for FIV.There is no evidence that FIV is transmissible to any otherspecies, including humans (123, 126, 255). Viruses similar toFIV have been documented in several nondomestic felids suchas lions, panthers, and bobcats (124, 144, 166), but it is unlikelythat they contribute significantly to the circulation of FIV indomestic cats. Genomic divergences between FIV and lentivi-ruses isolated from panthers in Florida were so marked as tosuggest that the transmission of these viruses between felinespecies is infrequent or nonexistent (166).The precise modes of FIV spread among domestic cats in
the natural setting are not yet clear. There is, however, strongcircumstantial evidence that biting is by far the predominantmeans of transmission. The highest prevalences of infectionhave invariably been found in feral colonies and in adulttomcats with unrestricted outdoor activity, which frequentlyexhibit aggressive behavior. Also, the gender-related and geo-graphical variations in prevalence mentioned above underlinethe importance of this route of transmission since males havea greater propensity for biting than female cats and the highestprevalences are found in parts of the world where cats are leftfree to roam outside ad libitum. On the other hand, a provokedbiting experiment has provided convincing evidence that trans-mission by this route is highly effective (255). An epidemiolog-ical association between FIV and the feline syncytium-formingspumavirus, which is also mainly transmitted by bites, has beenobserved (9). A similar association of FIV with the much moreeasily communicable feline leukemia virus (FeLV) is lesscertain (9, 35, 98, 255).FIV has been demonstrated in the saliva of experimentally
and naturally infected cats (132, 135, 255, 259). Virus isolationwas considerably less frequent from saliva than from periph-
eral blood mononuclear cells (PBMC). However, PCR analysisdemonstrated the presence of both viral RNA and DNA insaliva, suggesting that this fluid contains virus-infected cells orcellular debris as well as free virus. Interestingly, PCR revealedvirus genomes in a higher number of saliva samples and withstronger signals than expected on the basis of the virusisolation data (135). Although anti-FIV antibody is present inthe saliva of infected cats (186), feline saliva did not inhibitFIV growth in vitro. As PCR detection of viral sequences doesnot prove that infectious virus is present in a sample, thediscrepancy between isolation and PCR data was taken toindicate that a large proportion of the virus present in saliva isnoninfectious (135). If confirmed, such a conclusion wouldhelp explain the low efficiency of FIV transmission in nature.Modes of natural FIV transmission other than bite wounds
would appear to be rather ineffective. Sexual transmission hasnever been clearly documented (90), while vertical transmis-sion has been reported only from queens experimentallyinfected during pregnancy. In the latter experiments it wasimpossible to establish whether transmission had occurred inutero (the epitheliochorial placenta of cats is a considerablebarrier to cross) or through colostrum, milk, and maternalgrooming (28, 126, 176, 240, 245, 255). Attempts to isolate thevirus from colostrum and milk have failed (28, 245, 259). FIVwas isolated from fetal tissues of embryos aborted by a cat inthe primary phase of infection (77).The risk of fomite transmission is negligible, as FIV is
unlikely to survive for long outside the body and environmentalcontamination has never been associated with significant con-tagion. Noninfected cats living in close contact with infectedcats usually remain seronegative, most likely because catsmaintained in stable households for prolonged periods rarelybite each other (240, 259). However, it has been reported thatgreater than 50% of the seronegative cats housed for severalyears with infected companions contained FIV DNA, demon-strable by PCR and in situ hybridization, in their PBMC andbone marrow. The FIV DNA could be transferred to naiveanimals with blood and bone marrow, but the transfused catsalso remained free of recoverable infectious virus, antiviralantibody, and disease (40). We have confirmed these findingsat least partially (17). The existence of genome-positive butinfectious virus- and antibody-negative hosts has been de-scribed also for HIV and simian immunodeficiency virus (SIV);however, these findings remain controversial (113, 199). Onthe basis of recent data obtained from noninfected individualsexposed to HIV (199), it would be of interest to evaluate theanti-FIV cell-mediated immune status in seronegative FIVDNA-positive cats.Experimental transmission is easy by any parenteral route,
whether cell-associated or cell-free virus is used. Subcutaneousinoculation has been shown to be an efficient route of infectioneven in kittens with very high levels of maternal antibody,although scarification of the skin caused insufficient trauma toallow infection. In contrast, experimental transmission throughmucosal routes (vaginal, rectal, and with a lower efficiency,oral) required infected cells or large doses of cell-free virus(77, 145).
THE VIRUS AND ITS GENOME
Morphology and Biochemistry of the Virion
The morphology of FIV has not been studied in great detail.The mature extracellular virion is spherical to ellipsoid, 100 to125 nm in diameter, and bordered by an outer envelope withpoorly defined short projections or knobs. The elongated core
is composed of a conical shell that surrounds an eccentricelectron-dense nucleoid. A polygonal electron-lucent halo isoften visible between the core and a granular layer found justinside the envelope (141, 174, 228, 258). Thus, the electronmicroscopic appearance of FIV seems to be indistinguishablefrom that of other lentiviruses. FIV is rapidly inactivated bychemical disinfectants such as chlorine, quaternary ammonium
compounds, phenolic compounds, and ethanol and by heatingat 608C for few minutes (172).Figure 1 shows the presumed location in the FIV particle of
the envelope (Env) glycoproteins and the internal protein(Gag) components that have been identified with biochemicalor immunological methods or predicted from the study of theviral genome (see below). Each virion also contains two copiesof the positive-stranded, polyadenylated RNA genome (about9,200 bases long) and unknown numbers of molecules of thePol proteins reverse transcriptase (RT) and dUTPase (DU).As is typical of retroviruses, in the FIV virion Gag proteins areabout 20-fold more abundant than Pol proteins (55).Env glycoproteins of retroviruses are of importance because
they (i) are involved in receptor interactions with cells, thusdetermining the cell receptor-mediated tropism of the virus;(ii) exert membrane fusion activity (involved in virus penetra-tion and syncytium formation); and (iii) are the primary targetsfor antibodies and other effectors of immune functions. In thecase of FIV, the surface (SU) protein is heavily glycosylatedand has an apparent Mr of 95,000 (gp95), while the transmem-brane (TM) glycoprotein is less glycosylated and has an Mr of40,000 (gp40) (Fig. 2). Their amino acid compositions havebeen deduced from the nucleotide sequences of a number ofisolates (see below). Such studies have shown that, as in otherlentiviruses, the amino acid sequences of Env proteins arehighly variable, with up to 20% diversity between isolates. Thedifferences are mainly due to substitutions (deletions andinsertions would appear to be less frequent than in HIV), areconcentrated in specific segments of the molecule (variable [or
FIG. 1. Molecular anatomy of the FIV virion. Evidence for the presence ofthe integrase (IN) in the virion is still lacking. MA, matrix; NC, nucleocapsid. Forabbreviations of the other viral proteins, see text.
FIG. 2. env gene of FIV (A) and schematic structural models of the SU (B) and TM (C) glycoproteins of FIV (modified from reference 170 with permission of thepublishers). Shaded segments in panel A represent the variable regions. Shaded boxes in panels B and C indicate the amino acid sequences that have been shown tobe recognized by FIV-infected cat sera. The percentages of infected cat sera that reacted with the different regions are also given (references are given in parentheses).
V] regions), and generally do not affect glycosylation sites andcysteine residues thought to be crucial for correct folding ofthe protein. As in other lentiviruses, the V regions of FIV Envproteins (divergences in amino acid sequence of greater than10%) are interspersed within conserved sequences, designatedconstant regions (151, 170, 178, 197) (Fig. 3). Schematicstructural models of the spatial folding of the SU and TMglycoproteins have recently been built using predictive algo-rithms for secondary structures. Interestingly, the resultingmodels showed considerable similarities to the correspondingmolecules of HIV and other lentiviruses (Fig. 2), leading to thesuggestion that Env proteins of lentiviruses rely on a conservedstructural framework that tolerates extensive variations inamino acid positions (170).The major internal components of the virion are the matrix
protein of approximately 14.5 kDa, which has recently beenshown to be myristoylated (55); the capsid (CA) protein ofapproximately 24.5 kDa; and the nucleocapsid protein ofapproximately 7 kDa, which contains motifs characteristic ofretroviral nucleic acid binding proteins and is believed to beassociated with the genomic RNA to form the ribonucleopro-tein (50, 216). The relatedness of the predicted gag geneproduct of FIV to that of other lentiviruses ranges from 49%amino acid identity with equine infectious anemia virus(EIAV) to 40% with HIV type 1 (HIV-1) (50, 55, 165).The purified RT of FIV consists of a single polypeptide of
67,000 Mr which is nearly identical to that of HIV RT intemplate preference, divalent metal ion requirements, andsensitivity to active forms of competitive and noncompetitiveinhibitors (156–158, 196) although considerably different fromit in the predicted primary sequence (almost 60% divergence[221]). Also, the RNase H activity associated with FIV RTdisplays ion requirements, substrate preferences, catalytic ac-tivity, and sensitivity to polyanionic inhibitors very similar tothose of the RNase H associated with HIV RT (38, 73).Approximately one-half of the RT present in the virions lacksthe carboxy-terminal RNase H domain (55). The RT of FIV isbelieved to be as highly error prone as the correspondingenzyme of HIV, which is considered an important cause of thehigh genome variability observed in these viruses.The presence of a virus-encoded DU in FIV, EIAV, and
other retroviruses has been recognized recently by Elder andcoworkers (54). The enzyme has an Mr of 14,000 and anabsolute preference for dUTP. Its function is unknown, but it
does not seem to be essential, since several retrovirusesincluding the primate lentiviruses lack DU activity. It has beenspeculated that by keeping intracellular dUTP levels lowduring reverse transcription, the potential misincorporation ofuracil into viral DNA might be decreased. This function mightbe especially important during virus replication in cell typesthat have inherently low levels of cellular DU activity, such asthe noncycling macrophages. DU-minus mutants of FIV ob-tained by insertional mutagenesis replicated more slowly thanwild-type virus in transfected PBMC and not at all in primarymacrophages (244).
Genome Organization and Expression
The first isolate of FIV obtained by Pedersen et al. (174) wasalso the first strain to be characterized at the nucleotide level(164, 221). Since then, the genomes of a number of FIV strainsisolated in different regions of the world have been sequencedcompletely or partially. FIV diverges from other lentivirusesthroughout the genome. Its sequence organization, however, issimilar in complexity to that of other lentiviruses, with threelarge open reading frames (ORFs), gag, pol, and env, encodingthe internal structural proteins, the RT and other viral en-zymes, and the envelope proteins, respectively, as well asvarious small ORFs encoding regulatory proteins (Fig. 3). Thestructural features of the genome and phylogenetic treesconstructed following alignment of the Gag and Pol proteinsindicated that FIV is more closely related to the nonprimatelentiviruses EIAV, caprine arthritis encephalitis virus, andvisna maedi virus than to primate lentiviruses (129, 165, 178,221).Similar to all retroviruses, the provirus contains two long
terminal repeat (LTR) elements, one at each end, whichaccommodate multiple regulatory sequences. Upstream of the39 LTR is a polypurine stretch and downstream of the 59 LTRis a tRNA-Lys primer binding site which are believed to beinvolved in the initiation of plus- and minus-strand DNAsynthesis, respectively (129). In the FIV proviruses sequencedso far (23), LTRs have been found to have a length of 355 to361 bp, similar to those of caprine arthritis encephalitis virus,visna maedi virus, and EIAV and shorter than those of primatelentiviruses (about 460 bp for HIV-1) (140, 165, 221). Like thevisna maedi virus LTR, but unlike the HIV LTR, FIV LTRsappear to be strong basal promoters at least in certain cell
FIG. 3. General organization of the proviral FIV genome. The small ORFs shown are those most conserved among FIV isolates sequenced so far.
types and poorly active in transactivation (142, 215). Regula-tory sequences present in the LTR of FIV include one or twoTATA boxes, a consensus polyadenylation signal, one or twosets of imperfect repeats, a CCATT promoter, and a variety ofenhancer or promoter protein-binding sites (AP-1, AP-4,ATF-1, EBP20, NF-kB, etc.). Such regulatory elements havenot been found in all of the FIV isolates sequenced, whichmight explain the biological differences observed among dif-ferent isolates (101, 102, 151, 164, 178, 215). For some suchelements, the activity in promoting viral replication has beendemonstrated by in vitro mutagenesis of molecular clones ofFIV (143, 215). In HIV and SIV, the R region of the LTRcontains a sequence, called Tat responsive element (TAR),which functions as a recognition region for the transactivatingprotein Tat and has a stem-loop structure that appears topermit the appropriate presentation of the Tat protein. Theexistence of a similar stem-loop has been described also in theR region of the FIV LTR, but its role, if any, in enhancing FIVreplication has yet to be determined (140, 150). In fact,although a tat-like sequence has been identified in the genome,evidence of the existence in FIV of a transcriptional transac-tivator is scant (see below).The gag gene, spanning from approximately base 600 to base
2,000, is initially translated to yield a precursor polyprotein ofabout 50 kDa in sodium dodecyl sulfate (SDS) gel that issubsequently cleaved to yield (amino to carboxy order) thefunctionally mature matrix, CA, and nucleocapsid proteins (50,216). Cleavage most likely is done by the viral protease (PR).Processing of the Gag 50-kDa precursor did not occur in insectcells infected with a baculovirus expressing the gag gene butnot the protease gene of FIV (150, 241). The precise molecularmass of the mature Gag proteins has been recently determinedby Elder et al. (55), who have also identified most of theproteolytic processing sites within the Gag polyprotein. Inparticular, the CA protein was found to be processed in amanner very similar to that of HIV.The enzyme cassette pol gene, which spans from around
bases 1,900 to 5,200 and thus overlaps the gag gene by 109nucleotides, is in a21 reading frame with respect to that of thegag gene and is translated as a Gag-Pol fusion polyproteinproduced by ribosomal frameshifting, as in other retroviruses.The overlapping region contains a frameshift signal sequence,GGGAAAC, located near the 39 end of the gag gene, which isfollowed by a downstream sequence with the potential to forma tertiary pseudoknot structure apparently essential for effi-cient frameshifting, at least in baculovirus cells (149, 150). Theratio of frameshift to readthrough products in this system isestimated to be approximately 30% (149), higher than isobserved in HIV. The Gag-Pol fusion precursor protein iseventually processed into the functionally mature enzymesduring virus assembly. The Pol portion of the precursorpolyprotein (about 1,120 amino acids) is eventually cleavedinto (in order of translation) PR, RT, DU, and an endonucle-ase or integrase, most likely due to autodigestion (54, 55). Thepredicted lengths of these enzymes are 123, 559, 130, andabout 278 amino acids, respectively (54). Most recently, themolecular mass of FIV PR was determined to be 14.3 kDa(55).The primary translation product of the env gene (which
spans from around bases 6,250 to 8,850) is a precursorglycoprotein of 145 to 150 kDa (gp145) that is rapidly reducedin size to a molecule of 130 kDa (gp130), presumably becauseof trimming of the terminal glucose residues in the roughendoplasmic reticulum. Glycosidase digestion of gp150 andgp130 or treatment of infected cells with tunicamycin resultedin proteins of 100 and 75 to 80 kDa, respectively, indicating
that both precursors are heavily glycosylated (218, 241). Thepattern of glycosylation may vary depending on the cell typeused for growing the virus (187). gp130 is subsequently pro-cessed into the mature glycoproteins SU (gp95) and TM(gp40) of the viral envelope by proteolytic cleavage in theGolgi complex (218). These structural proteins, however, donot exhaust the coding potential of the env gene (2,570 bp inthe Petaluma isolate). The gene also encodes a long amino-terminal presequence (149 amino acids) which also encodesthe first coding exon of the rev gene and a hydrophobic signalpeptide (20 amino acids) whose precise functions are not clearbut appear to be involved in the proper processing andtransport of the SU glycoprotein, from which they are cleavedduring maturation (217).The sequence of the env gene is the least conserved between
FIV isolates. As already alluded to in the description of themature Env proteins, nucleotide variations are not distributedrandomly throughout the env gene but cluster in specificregions (Fig. 3 and 4). The number and precise localization ofV domains are still under definition. Pancino et al. (170)defined nine V regions, two of which (V1 and V2) are locatedin the amino-terminal presequence and signal regions; four(V3 to V6), in the SU protein; and three (V7 to V9), in the TMprotein (Fig. 2). This classification is largely accepted and willbe used in this article. Analysis of the rates of silent andreplacement substitutions in the variable regions has led toindirect evidence for positive selection of variants, possiblyexerted by immune pressure. The rate of change in the envgene has been estimated to be about 3 3 1023 nucleotidesubstitutions per site per year (70). Phylogenetic analysis of thenucleotide sequences encompassing the V3, V4, and V5 re-gions showed that the degree of divergence among isolates is afunction of geographical distance and segregation, reminiscentof HIV-1 subgrouping (Fig. 5) (23, 197, 212). According to arecent study focusing on a 648-nucleotide sequence encom-passing regions V3 to V5 of the env gene, by criteria similar tothose used to group HIV, isolates of FIV could be divided intoat least three distinct sequence subtypes. Pairwise sequencedivergence ranged from 18 to 26% between subtypes and from3 to 15% within each subtype and was as high as 4% withingenomes derived from individual infected cats (212). Furtherevidence for geographical clustering of genetic variants of FIVcomes from recent comparisons of gag and pol sequences ofAustralian and American isolates (69). The gag and pol genesare more conserved than the env gene among FIV strains;however, the amino acid sequence of the CA protein of theJapanese isolate TM2 and the Italian isolate Pisa-M2 diverge5% (129) and 7% (17), respectively, from that of the Petalumaisolate, and the amino acid sequence of the nucleocapsidprotein was found to diverge as much as 20% in the TM2isolate (129). The biological significance, if any, of thesesubtypes in terms of transmissibility, pathogenicity, and vaccinestrategies is yet to be understood.The only two regulatory genes of FIV that have been
characterized to some extent are rev and vif. Proviruses of HIVand SIV lacking a functional rev gene cannot produce Gag, Pol,and Env proteins and are replication defective. In FIV, Revactivity has been mapped to two areas: the first exon at theintergenic region from the 39 end of the pol gene through theL region of the env gene, and the second exon to the 39 end ofthe env gene (177, 233). By immunofluorescence, the 23-kDaRev protein was localized in the nucleus of infected cells. TheRev protein binds to an element, called the rev-responsiveelement, that in FIV has been mapped to a 243-bp fragment atthe end of the env ORF and partially overlaps at the 39 LTRand not at the SU-TM protein junction as in other retroviruses
(Fig. 3) (177). In HIV the Rev protein is believed to promotetransport of unspliced viral mRNA containing rev-responsiveelements from the nucleus to the cytoplasm and also appearsto increase the stability of unspliced and singly spliced viralmRNA species containing rev-responsive elements. A recentinvestigation revealed that, in addition to the genomic FIVRNA, FIV-infected cells contain at least five short FIV-specifictranscripts and that the smaller ones are doubly or triply
spliced mRNA (233). Thus, the transcription of the FIVgenome is as complex as that of other lentiviruses withelaborate alternative splicing, which suggests the existence ofmultiple levels of control. Experiments with rev-minus mutantsof FIV demonstrated that in the absence of this regulatoryfactor, no unspliced or singly spliced viral mRNA is expressedand virus production is abolished or diminished (104, 177,233). Further studies will be necessary to completely under-stand the role of FIV Rev in modulating FIV replication.The vif gene (or ORF-1) of FIV is similar in size and
location to the vif gene of primate lentiviruses (178, 221).vif-minus molecular clones of the isolate TM2, once trans-fected into cells of the fibroblastoid Crandell feline kidney(CrFK) line, replicated in a manner indistinguishable from thatof the wild type (232), but infection with the vif-minus mutantwas possible only by cocultivation with infected cells and notwith cell-free virus, thus reproducing a biological property oflentiviral vif-minus mutants. The molecular basis of the activityof this gene is not known.Some other small ORFs present in the FIV genome are
conserved in different isolates, suggesting that they performimportant functions. However, their functions and RNA tran-scripts are still poorly understood. Nucleotide homologysearches have revealed no obvious similarities between theseORFs and the regulatory gene sequences of HIV and SIV (23).One such ORF has been partially characterized by Tomonagaet al. (231). ORF-A (or ORF-2) is similar in location (betweenthe vif and rev genes) (Fig. 3) and size to the first exon of thetat gene in primate lentiviruses. An ORF-A mutant of the TM2clone retained the capacity to replicate in established CD41
T-cell lines, but the resulting viral progeny replicated poorly,especially in fresh PBMC. Although the exact role of ORF-Ain the life cycle of FIV remained unclear, the results wereconsistent with the possibility that its product is a transactiva-tor. Since infection of primary PBMC is presumably closer to
FIG. 4. Predicted amino acid sequences of the V3 and V4 regions of FIV SU glycoprotein of 16 FIV isolates. The hatched box indicates the domain of the V3 regioninvolved in antibody-mediated neutralization of FIV infectivity for CrFK cells (120).
FIG. 5. Degree of relatedness among 17 FIV isolates. The maximum-likelihood phylogenetic tree based on 481-bp sequences of the env gene is shown(reproduced from reference 197 with permission of the publisher).
what occurs during a normal infection in cats, it is likely thatORF-A mutant viruses replicate poorly in vivo.
Replication and Effects on Cells In Vitro
In vitro, FIV is tropic for cat T lymphocytes but also infectscells of the monocyte-macrophage-microglia lineage and astro-cytes (27, 45, 172). The similar cell tropism of HIV, and themarked reduction of circulating CD41 T lymphocytes observedin FIV-infected cats, led to the initial suggestion that, as forHIV, the cat CD4 antigen might be the cellular receptor forFIV. This assumption, however, was soon discarded as subse-quent observations showed a lack of correlation between theexpression of CD4 on cells and their ability to support FIVreplication (26, 250). In fact, the spectrum of cultured lym-phoid cell types that have been shown to support FIV growthupon in vitro infection is wide and includes primary cultures ofCD41 and CD81, null T cells, and a variety of interleukin-2(IL-2)-dependent and -independent, FeLV-transformed orFeLV-free T-cell lines with different surface phenotypes (56,250, 259). The conclusion that the CD4 molecule was notinvolved in FIV entry (at least in vitro) was strengthened by theobservation that transfection and stable expression of clonedcat CD4 on nonlymphoid feline cells failed to render thesecells susceptible to the highly lymphotropic TM1 and TM2isolates of FIV (155). It is noteworthy that, despite this majordifference from HIV, FIV infection produces the same effectsas HIV infection on CD41 lymphocytes both in vitro and invivo.Recently, Hosie et al. (91) identified a putative non-CD4
receptor by means of a cell-specific antibody designated vpg15which, although unable to bind the virus, efficiently blocks FIVinfection of susceptible cells in vitro. The vpg15 ligand hasbeen characterized as a 24-kDa protein whose expression isconsistently decreased when cells become FIV infected. On thebasis of reactivity with an anti-human antibody that cross-reacts with feline cells, this molecule has been tentativelyidentified as the feline homolog of human CD9, a markerexpressed on a range of cell types of both hematopoietic andnonhematopoietic origin (251). Alternative routes of entry intocells are possible but have not been demonstrated. Underlaboratory conditions, low concentrations of antibody enhanceHIV and SIV replication due to facilitated entry into cells viaFc and complement receptors (146). In vitro enhancement ofFIV by subneutralizing concentrations of antibody has beenobserved, but the receptors involved have yet to be determined(7).Different FIV isolates may differ in their in vitro host range
even with respect to cells apparently showing the same surfacephenotype (100, 178, 232). The reasons are not known butmight involve differences in the Env protein as well as in thecomposition of the proviral LTR. Miyazawa et al. (142)showed that following transfection of an infectious molecularclone, FIV replicated in some but not all of the otherwisenonsusceptible simian and human cells tested and that in theresistant cells the FIV LTR displayed a much lower level ofbasal promoter activity than in cells that replicated the virus.Some FIV strains (PPR, TM1, and TM2) could be adapted togrow in CrFK cells, but others could not (141, 178). Inability toinfect these cells seems to be a stable trait as it was conservedin the progeny of infectious molecular clones transfected intoCrFK cells (178). Although CrFK cells express the putativevpg15 receptor marker (91), the latter experiment and otherdata suggest that failure to replicate in CrFK cells is not due toa postentry block (129, 140, 142, 207, 209). It should be noted,however, that the Italian isolate Pisa-M2 could be adapted to
grow in CrFK cells only after several passages in the lymphoidcell line MBM (7). This indicates that the passage history ofthe virus may affect tropism, possibly because of differences insurface proteins acquired by the virus from the membrane ofthe host cell (168). Differences in cell tropism of FIV isolatesin vivo have also been noted (236).The consequence of FIV infection on cells may vary depend-
ing on both the viral isolate and the host cell (148, 230). Inlymphoid cells, cytopathic effects (CPE) usually become evi-dent 7 to 14 days after infection and include cell ballooning,syncytium formation, and cell lysis (174). Willett et al. (250)observed that syncytium formation coincided with a sharp risein virus production. However, formation of syncytia is pro-nounced in some cells but is not a general occurrence. Forexample, syncytia are very evident in CrFK cells adapted togrow in a medium containing a low concentration of serum(237) (Fig. 6) but were not noted in infected CD31 CD4282
lymphoid MBM cells (136). During early infection of a T-cellline, cellular death was preceded by ballooning and syncytiumformation, but in long-term infected cultures only cell lysisoccurred (250). Syncytia were also formed when human lym-phoblastoid cells were exposed to FIV-infected cells or toconcentrated FIV. This effect occurred in the absence of viralreplication, suggesting that it was produced from the viralmaterial added to the cultures (229). In a study on thesusceptibility of feline neural cells to FIV, Dow et al. (45)noted that astrocytes were the most susceptible and underwentsyncytium formation and death, microglial cells productivelyreplicated the virus without obvious CPE, and endothelial cellsproduced little virus with no CPE. Importantly, oligodendro-cytes and neurons remained free of detectable infection andCPE. There are also preliminary indications that FIV isolatescan be grouped as syncytium forming and nonsyncytium form-ing (65). A recent examination of genetically constructedchimeric viruses indicates that the env region of FIV containsdeterminants of CrFK tropism but that other determinantsmay also be involved in tropism (209). Interestingly, in culturesof CD41 cells, expression of CD4 antigen was reduced by FIVinfection in a fashion similar to that observed with HIV (230,250), but CD4 levels returned to normal in chronically infectedcells (250). In the case of FIV, such down-regulation cannot beascribed to interference due to intracellular binding of theCD4 protein by the viral SU protein as it can be in the case ofHIV (113).A number of cell lines persistently infected with FIV have
been established. These include CrFK cells (259) and severallymphoid cell lines (141, 250, 254). For example, Yamamoto etal. (257), by cloning from an IL-2-dependent culture of mixedPBMC of an FIV-infected cat, have developed an IL-2-independent CD46 CD81 T-cell line (FL-4 cells) that pro-duces large quantities of FIV and is presently used in manylaboratories as a source of virus for experimental work anddiagnostic assays. It should be noted, however, that the FIVproduced by persistently infected cells may be consistently lessinfectious and virulent and induce more feeble antibody re-sponses than non-cell culture-adapted virus (11). The CPEobserved in persistently infected cell lines is low to minimal,although the virus produced retains its CPE capabilities fornewly infected cells. However, cell behavior can be affected, asshown by changes in surface marker expression (250) and bythe finding that treatment with tumor necrosis factor alpha ofpersistently infected CrFK cells, but not control cells, led tocellular changes characteristic of apoptosis (162). The mecha-nisms responsible for these phenomena were not examined indetail, but it is of interest that similar changes have beenobserved in cells persistently infected with HIV.
FIV interactions with macrophages have not been exten-sively studied. However, there are indications that the virustends to establish a latent infection in these cells. Freshlyharvested peritoneal macrophages were readily infected invitro, but after an initial burst of viral replication, the RTproduced progressively decreased, although there was a con-comitant marked increase in number and size of the syncytiaproduced (27). In addition, cultures of peritoneal macrophagesderived from long-term infected cats expressed little or no RTactivity. However, the virus could be reactivated by stimulationwith a phorbol ester and cocultivation with fresh PBMC (27,128). The ability of the viral genetic information to persist incells without active virus replication might account for the longlatent period of FIV disease.Intracellularly, the replication of FIV is assumed to proceed
along pathways similar to those of the much better understoodHIV. Information, however, is scanty. As initiation of tran-scription is dependent on both viral and cellular regulatoryproteins, the rate of viral transcription is markedly affected bythe rate of transcription of the infected cell (101, 102, 215).This can explain why T-cell activation is a prerequisite forproductive infection of PBMC in vitro. Studies with baculovi-rus vectors showed that the unprocessed Gag precursor canassemble, most likely following appropriate myristoylation,into spherical viruslike particles at the cellular membrane andbe released as such, without the need for other viral products,while inclusion of the intact PR gene into the gag-expressingvector resulted in efficient processing of the 50-kDa polypro-tein into mature Gag proteins and in the failure to produceviruslike particles. These results led to the suggestion that FIVGag protein cleavage occurs during or after budding of thevirions (150). Following synthesis and processing, the viralenvelope glycoproteins are eventually deposited at the surfaceof the cell (217). Then, the complex of capsid proteins,replicative enzymes, and viral RNA assembles into a closed
spherical particle that buds from the plasma membrane ofinfected lymphocytes. During this process, the buds have thecrescent-shaped appearance of other lentiviruses (258).
THE INFECTION
Demonstration of FIV Infection
Current methods for detecting FIV infection in cats includevirus isolation, immunological tests for virus-specific antibod-ies or antigens, and molecular tests for FIV genetic sequences.No diagnosis can be made on clinical grounds alone.By contrast with most other viruses, because of the ability of
lentiviruses to persist, an unambiguous demonstration of anti-viral antibodies in serum can be considered adequate evidencefor ongoing infection. Serological assays for FIV antibodies arerapid and sensitive, and though they provide only indirectmarkers of infection, they are usually sufficient in veterinarypractice for diagnosis of infection. A variety of assays havebeen described but enzyme-linked immunosorbent assays(ELISAs) using cell culture-derived, gradient-purified virus orrecombinant viral core proteins as antigen are by far the mostcommonly used. Easy-to-use ELISA kits for the detection ofanti-FIV antibody are commercially available. The sensitivityand specificity of such assays are usually satisfactory, butfalse-positive and false-negative results have been reported.False-positive results may arise, especially when the assays areperformed with whole-virus lysates and the serum samples arefrom cats vaccinated for other pathogens which may thereforepossess antibody to cell culture components (87, 139, 160, 172,189, 211, 241, 255). Indirect immunofluorescence assays arealso used but, because they are considerably less specific,should be confined to use for screening purposes.When needed, ELISA results can be confirmed by Western
immunoblotting (WB), radioimmunoprecipitation followed by
FIG. 6. Syncytium formation by FIV. (A) Syncytia produced in CrFK cells adapted to grow in medium with low serum concentration (stained with 0.3% crystal violetin 20% methanol; magnification, 380). (B) FIV antigen expression by the same syncytia, demonstrated by indirect immunofluorescence with a polyclonal cat serum(magnification, 3160).
SDS-polyacrylamide gel electrophoresis analysis, or neutraliza-tion (51, 53, 87, 238, 255). These tests, however, should not beregarded as completely error proof (172). A recently devel-oped competitive ELISA using CA antigen-specific monoclo-nal antibodies also proved useful in eliminating false-positiveresults (122). WB assays detect antibodies against individualvirus proteins and may also be important for studies concern-ing FIV pathogenesis. Radioimmunoprecipitation also detectsantibody that remains occult by immunoblotting (87). Immu-noblotting and radioimmunoprecipitation studies have shownthat several virion and precursor proteins and glycoproteins ofFIV are immunogenic in infected cats (50, 216). All sera ofnaturally FIV-infected cats recognized the Env proteins. Mostsera also recognized the CA protein and one or more addi-tional Gag proteins, but a few serum samples showed antibod-ies only to Env proteins (51, 66). Antibodies to the Envglycoproteins, however, are usually undetectable by WB whenpurified virus is used (160), most likely because such moleculesare easily sheared from the virion during purification andinefficiently transferred to the nitrocellulose sheets (51, 87).For this reason, WB methods using FIV-infected cell lysateshave been proposed. The interpretation of WB results may bedifficult, especially when the strips are prepared with infectedcells or poor-quality virus, since protein bands of cellular originmay be formed in positions only slightly different from those ofviral proteins. O’Connor et al. (160) regarded as positive thepresence of reactivity to two core proteins, while according toHosie and Jarrett (87), the minimum requirement for apositive WB was the detection of antibody to the SU glycop-rotein and to at least three core proteins. Some of theserological assays described above were found to be goodpredictors of infection when applied to saliva as well (186).Because saliva in cats is much easier to collect than blood, theperformance of saliva-based assays in the field deserves carefulevaluation.The identification of conserved immunodominant B-cell
epitopes in the SU and TM proteins (see below) has allowedthe development of a peptide-based ELISA and immunoblot-ting tests that can be useful for diagnostic purposes (6, 42, 64,120). Recently, Verschoor et al. (241), using a recombinantpeptide ELISA, found that some infected cats may lack orpossess very low levels of antibody to either Gag or Envproteins, thus emphasizing the necessity of testing for antibod-ies to both viral structures.FIV isolation is too laborious and time-consuming for
routine use and requires well-equipped laboratories andtrained personnel; therefore, it is reserved for research pur-poses. Reportedly, the virus has been isolated from cats that donot show detectable antibody (85). The virus is usually isolatedby coculturing PBMC of infected cats either with primary blastcells obtained by mitogen prestimulation of lymphocytes fromnaive, specific-pathogen-free (SPF) cats or with cells of certainfeline T-lymphoid lines that may be at least as permissive asfresh primary blast cells (65, 136). The cocultures are usuallymonitored for Mg21-dependent RT and CA antigen produc-tion over periods of 5 to 7 weeks. They tend to become viruspositive considerably earlier when clinical specimens are de-rived from diseased cats rather than from asymptomatic ani-mals (9, 135). By using similar procedures, the virus has alsobeen isolated from several sources other than PBMC (seebelow). Quantitative coculture techniques that use gradednumbers of infected cells have also been used (136, 138). Apoint usually not fully appreciated is the ease with whichcross-contamination can occur during the long cultivationperiods needed for FIV isolation. We encountered this prob-lem when a lentivirus was isolated from an FIV-seropositive
sick lion (183). When the gag gene of this isolate was se-quenced, it was recognized as a strain of FIV that is routinelypropagated in our laboratory (17).Several capture ELISA methods for detecting FIV CA
antigen have been developed (40, 122, 211, 227) and areextensively used to monitor virus growth in cell cultures. Theycan be rendered quantitative by comparing the readings givenby the sample with a standard curve calibrated for viral CAantigen. One such method, based on the use of noncompetingmonoclonal antibodies and found to detect as little as 0.25 ngof CA protein per ml, has been validated also for detectingantigenemia in infected cats. For optimum sensitivity, acidtreatment of sample sera was required to disrupt immunecomplexes before testing. With this procedure CA antigen wasdetected in 46% of infected cat sera tested (122). The availabledata are still too limited to predict whether the measurementof antigenemia will become a routine prognostic marker inFIV infection as it is in human AIDS (239). FIV antigens alsomay be searched for in infected tissues by using immunohisto-chemistry methods (47, 122, 133, 236).So far, FIV genome detection by in situ hybridization has
had limited use (14, 15, 40). Instead, viral DNA and RNAamplification by one or two-step PCR methods has beenextensively applied to detect the viral genomes in infectedtissues and cell cultures. The primers chosen are located inconserved sequences of the gag, pol, and, less frequently, theenv regions (40, 84, 135, 175, 198). Discrepancies between PCRand isolation or serological results have been described andmay stem from a variety of factors, including insufficientnumbers of infected cells in the sample and presence ofdefective integrated proviruses in the cells. PCR methods alsocarry the threat of false-positive results caused by unrecog-nized carryover of amplified DNA into experimental samplesand other artifacts. The use of multiple sets of primers fromdifferent regions of the viral genome and other precautions aswell as cautious interpretation of the results are mandatory.There is a great need for simple and reliable quantitative PCRmethods for staging the infection and monitoring FIV levels incats; a limiting cell dilution method (56), a method using anexternal standard (179), and one using a deleted internalstandard (180) have been described recently but have yet to befully evaluated.
Host Invasion and Persistence
Little is presently known about where primary FIV replica-tion takes place after its entry in the host, the routes it uses todisseminate within the body, and the time needed for theinfection to become systemic. Matteucci et al. (135) isolatedFIV from virtually all lymphoid tissues examined as early as 1week postinfection (p.i.). At this time, salivary glands were theonly isolation-positive nonlymphoid tissue. By 3 weeks p.i.,however, all nonlymphoid tissues examined, including thebrain, yielded positive cultures, while at later times virusisolation from such tissues was less consistent than fromlymphoid tissues. Whether and to what extent isolation fromtissues reflected local viral replication or the presence ofcirculating infected PBMC and plasma were not determined.Virus is readily isolated from the PBMC and plasma ofexperimentally infected cats shortly p.i. (135). Essentially sim-ilar results have been reported by Dua et al. (47). We currentlyuse plasma of cats infected for 2 weeks as a source of virus forin vivo infection; the titers are approximately 103 50% catinfectious doses per ml. Beebe et al. (14) confirmed by in situhybridization that lymphoid tissues (lymph nodes, tonsils,thymus, spleen, and gut-associated lymphoid tissue) are the
primary early targets of FIV. In the first week p.i., the highestconcentrations of viral DNA and RNA were detected in thebone marrow, thymus, and lymph nodes and lower levels werefound in the other tissues examined, including kidney, lung,and liver. The lymph nodes contained viral genomes mainly inthe germinal centers early p.i. and mainly in the extrafollicularareas at later stages. Viral RNA was found in the nonlymphoidtissues solely at later stages of infection and appeared to bepresent mainly in the interstitium, especially around ducts, inblood vessels, or in inflammatory lesions. Cells other thaninterstitial mononuclear leukocytes or mononuclear inflamma-tory cells did not appear to be infected (14).As discussed, FIV replicates in a wide spectrum of cultured
cell types. This, however, does not necessarily reflect thesituation in vivo, as it is well known that cell permissiveness toviruses can be modified substantially by culture. Knowledgeabout the cell types that support FIV replication in vivo islimited. Ex vivo, FIV has been detected in lymphocytes (174),peritoneal macrophages (27), and renal (14, 184, 259) andcentral nervous system (CNS) (14, 46) cells of infected cats. Byimmunohistochemistry, lymph nodes were shown to containCA antigen in small lymphocytes, follicular dendritic cells, and(especially during late stages of infection) macrophages (122,133, 236). According to Toyosaki et al. (236), follicular den-dritic cells might represent an important target for FIVreplication and persistence, although variation in the involve-ment of these cells by different virus isolates was noted. Bydouble-staining methods, viral antigen-positive lymphocyteswere mainly CD41 T cells. Most notably, no viral antigen-expressing CD81 T cells were seen. Beebe et al. (14) foundthat in the first days p.i. 25 to 75% of the virus-positive cells inmost tissues were T lymphocytes. The situation, however,changed abruptly when clinical signs of primary infectionbecame evident, as a dramatic increase in the proportion ofFIV-infected macrophages and other non-T cells was noted.Recently, English et al. (56) performed an accurate PCR
analysis of lymphocyte subsets sorted from peripheral blood ofinfected cats. The provirus was detected in CD41, CD81, andnull T lymphocytes and also in a population of immunoglob-ulin-positive (Ig1) cells which did not appear to be monocytesand had several markers of B cells. At 2 to 6 weeks p.i., thehighest level of provirus was found in CD41 cells, while fromthe third month onwards the highest proviral load was found inthe Ig1 cells (positive signal given by 103 cells). Once culturedin vitro, all three cell populations obtained from infected catsproduced virus. In contrast, little or no provirus was found inmonocytes. As it was impossible to exclude that part of theprovirus signal observed was generated by Ig-coated nulllymphocytes or other cells, the authors interpreted thesefindings with great caution. The presence of abundant FIVprovirus in B lymphocytes would undoubtedly represent amajor deviation from what is known about the cell tropism ofother lentiviruses.The vast majority of bone marrow cells apparently remain
virus-free throughout the course of infection. By in situhybridization, FIV RNA was detected only in limited numbersof megakaryocytes and unidentified mononuclear cells ofmorphologically abnormal marrows (15). The latter cells werealso positive for viral CA antigen (122). Systematic studies ofthe virus burdens found in cats at different stages of infectionhave yet to be performed. The viral loads found in the PBMCof 11 asymptomatic cats, though extremely variable, were oftenlower than those detected in 12 symptomatic animals (136).This finding might indicate that the numbers of provirus-carrying PBMC increase with progression to disease.Once established, lentiviral infections usually persist for the
lifetime of the infected hosts. FIV is no exception. Althoughanecdotal cases of cats developing a transient viremia notfollowed by seroconversion or other signs of infection exist, insome studies virus was isolated from 100% of seropositive cats(64, 85). In other studies a single cultural attempt sufficed toisolate the virus from over 80% of naturally infected cats (135).Rare cases of truly seropositive cats have, however, failed toyield positive cultures on repeated testing (17). The mecha-nisms of FIV persistence in spite of the prompt, intense, andsustained immune responses developed by infected cats havenot been investigated. It is likely that several mechanismsamong those generally invoked to explain viral persistence areinvolved, including virus ability to hide in immunocompetentcells, low sensitivity of the virus to serum-mediated neutraliza-tion, and antigenic drift. As already discussed, latent infectionof macrophages has been described (27) and variability of FIVsurface molecules is high. That FIV sensitivity to antibody-mediated neutralization is low will be discussed below.
IMMUNE RESPONSES TO THE VIRUS
Immune responses to FIV are still under scrutiny. Whilecell-mediated responses are still essentially uncharacterized,antibody responses are prompt and vigorous and remainelevated during the long course of infection. The immuneresponse probably contributes to keeping the infection undercheck, as suggested by the prolonged clinical latency, but failsto entirely eliminate the virus. Whether such acquired immu-nity is protective against reinfection by different strains of FIVis not known. It is assumed that the persisting virus leads to aprogressive derangement of immune functions which eventu-ally renders the host defenseless and the infection ultimatelyfatal. As brilliantly suggested for HIV (247), FIV mighteventually become its own opportunistic infection.
Antigenic Determinants of FIV
Characterization of FIV antigens is still in its infancy.Because of their potential importance as protective antigens,attention has been mainly focused on envelope glycoproteins.Structural modeling studies based on published FIV nucle-otide sequences and screening by ELISA and Pepscan ofselected synthetic and recombinant peptides have led to therecognition of several antibody-reactive regions on the SU andTM glycoproteins.By screening a random peptide library, Pancino et al. (169)
identified five regions of the SU glycoprotein which wererecognized by variable proportions of infected cat sera; oneregion appeared to be isolate specific, while the others weremore or less extensively broadly reactive. Analysis by Pepscan(Cambridge Research Biochemical, Cambridge, United King-dom) of the peptide most frequently recognized (amino acidpositions 388 to 424) led to the identification of a smallersegment mainly located in the V3 region of the molecule(amino acid positions 398 to 408) which was recognized by100% of naturally infected cat sera and 88% of sera fromexperimentally infected animals (6). Lombardi et al. (120),using partially overlapping synthetic peptides, and deRondeet al. (42), studying a series of large recombinant peptides,came to similar conclusions. Another important immuno-dominant domain has been localized at the carboxy terminusof the SU molecule (5, 42, 169).On the TM glycoprotein a sequence containing a cysteine
loop exposed at the apex of the molecule, which in variouslentiviruses including HIV-1 and HIV-2 represents a con-served structural motif and contains an immunodominant
region, was recognized by 100% of sera from infected cats.Antibodies to this region of FIV developed within a few weeksp.i. and were maintained at high levels for extended periods (6,42, 64). The pentapeptide QNQFF, which is conserved amongthe several FIV isolates sequenced, was the minimal sequencerequired for reactivity (6). Another peptide located closer tothe carboxy terminus of the molecule was found to react with80% of sera from naturally infected cats (6). Other regionsseen by variable proportions of infected cat sera are shown inFig. 2. It is of interest that immunodominant sites tend tocoincide with variable regions. It has been speculated that thismight represent a means of diverting immunologic recognitionfrom conserved segments which are more likely to exertimportant viral functions (178).The V3 region of HIV-1 SU glycoprotein contains promi-
nent immunoreactive domains involved in virus neutralizationas well as important determinants of cell tropism, cytopathicity,and infectivity (33, 113, 153). Similar observations have beenmade for other lentiviruses (8). Reasoning that the V3 regionsof FIV and HIV-1 have a number of structural analogies,including a similar distance from the carboxy terminus of themolecule, the presence of a Cys-Cys bond, a computer-pre-dicted b-turn, and a glycosylation site, Lombardi et al. (120)focused on the V3 region of the FIV SU molecule. Analysis ofthis region by Pepscan and ELISA with four partially overlap-ping peptides revealed three linear antibody-binding domains,two of which mapped in the amino-terminal part of V3. Theimmune reactivity of the V3-based peptides was probably dueto the presence of both B- and T-cell epitopes because thepeptides not only reacted with infected cat sera but alsoelicited a delayed-type hypersensitivity reaction when injectedintradermally into FIV-infected cats (Table 1). This study alsoshowed that the V3 region of FIV contains a domain involvedin the neutralization of FIV infectivity for the fibroblastoidCrFK cells: the peptide GSWFRAISSWKQRNRWEWRPDF(Fig. 4) inhibited the FIV-neutralizing activity of pooledimmune cat sera, and cat sera raised against this peptideeffectively neutralized FIV infectivity for CrFK cells (120).This finding has been confirmed by characterization of aneutralizing serum produced by immunizing rabbits with a
recombinant peptide (42) and a neutralizing monoclonal anti-body which appeared, however, to bind a conformationalepitope (52). Thus, it would appear that the V3 region could beof great interest as a component of a candidate vaccine againstFIV.It should be mentioned, however, that similar to neutraliza-
tion of other lentiviruses (146, 153), neutralization of FIVappears to be a very complex phenomenon. Thus, Siebelink etal. (206, 210) showed that two different single amino acidsubstitutions in the V4 and V5 regions of the SU glycoprotein(amino acid positions 483 and 560, respectively) conferredresistance to virus neutralization. On the other hand, Baldi-notti et al. (7) found that sera of infected cats that exhibitedelevated titers of highly efficient, widely cross-reactive neutral-izing activity when assayed on CrFK cells exerted a much lowerneutralizing effect when assayed in lymphoid cells, which wasevident only with homologous virus. Also, at subneutralizingdilutions, sera of infected cats enhanced FIV production bylymphoid cells, a phenomenon not observed with CrFK cells.These quantitative and qualitative discrepancies were attrib-uted to differences in the passage histories of the viral prepa-rations and/or in the characteristics of the cells used, but suchpossibilities could not be tested because FIV adapted toproduce syncytia in CrFK cells proved poorly infectious forlymphoid cells and vice versa. It is worth noting that immunesera also inhibited the fusion of feline FIV-infected cells withhuman T-cell lymphotropic virus type 1-infected cells (222).As is typical of retroviruses, the gag proteins of FIV are
abundant in the virion and also highly immunogenic. Indeed,when mice were immunized with purified FIV, the bulk ofanti-FIV antibodies was directed against the CA protein.Epitope mapping of this molecule with murine monoclonalantibodies detected four distinct B-cell epitopes that provedimmunogenic also for the natural host (119). The RT is alsoimmunogenic, as determined by immunoblot analysis andinhibition of enzyme activity (59). Whether epitopes of theinternal virion proteins are expressed on the surface of FIV-infected cells is a matter of great interest for vaccine designand deserves further investigation (122, 134, 154).FIV shares important antigen reactivities with lentiviruses
recently described in several nondomestic felids (124). Incontrast, antigenic relatedness to other lentiviruses is low. Noserological cross-reactions were detected between FIV andprimate lentiviruses (165, 258), and only minor cross-reactivi-ties were seen between the major core proteins of FIV andnonprimate lentiviruses (50, 165, 216). Sera of horses infectedwith EIAV also recognized the SU glycoprotein of FIV, butthe immunoreactivity was dependent on glycosylation (216).
Antibody Responses
Though some late seroconversions are occasionally observed(259), after experimental exposure to the virus, serum antibod-ies to FIV usually become detectable within 2 to 6 weeksdepending on virus dose and other variables. Antibodiesdirected to the major core proteins and to envelope glycopro-teins develop at approximately the same time and tend toremain high throughout the lifespan of the infected animal (5,64, 120, 160, 258). Antibodies to RT were first detectedrelatively late in infection and increased over time up to 3 yearsp.i. (59). Further studies will be necessary to evaluate whetherconsistent differences exist in the temporal development ofantibodies to individual epitopes and to determine whetherthey can be used to stage the infection or predict prognosis(64). Vertically acquired antibodies found in kittens born ofFIV-infected queens decline to undetectable levels in 2 to 3
TABLE 1. Delayed-type hypersensitivity response to syntheticpeptides of the V3 region of FIV SU glycoprotein in experimentally
a From reference 17.b 2, negative skin test; 1, positive skin test.c SPF cats infected with FIV 5 to 12 months earlier were injected intradermally
with 50 mg of the indicated synthetic peptides derived from the V3 region of FIVas described in reference 120 or of an irrelevant control peptide and examinedfor delayed-type hypersensitivity skin reaction 48 to 72 h later.
months postdelivery, unless the kittens become infected andproduce an active response to the virus (28).The development of FIV neutralizing antibodies has been
studied only recently by examining the ability of cat sera toinhibit FIV replication (60, 61) and FIV-induced syncytiumformation in CrFK cells (237, 238). Tozzini et al. (238) foundthat active production of neutralizing antibodies was evident at5 to 6 weeks p.i., reached a plateau 3 to 4 months later, andpersisted throughout a prolonged observation period. Neutral-ization was potentiated by complement, an effect that was atleast partly attributed to virion lysis (60). As cat sera obtainedfrom field cats in different geographical areas effectively neu-tralized different virus isolates, these studies indicated that,like HIV, FIV possesses epitopes capable of inducing broadlyreactive neutralizing antibodies (238). As discussed above, onesuch domains maps in the V3 region of the SU glycoprotein(120).Whether the broadly reactive neutralizing antibodies de-
tected by assays performed with CrFK cells play any role inprotective immunity in vivo is unknown. Neutralizing antibodytiters were similar in symptomatic and asymptomatic infectedcats (135). In vitro exposure to anti-FIV antibody had littleeffect on the ability of FIV to infect cats: sera with high titersof antibodies that neutralized FIV for CrFK cells failed toinhibit the infectivity for cats of FIV injected intravenously orof FIV infectivity for lymphoid cells (7). In addition, thelikelihood of isolating FIV from saliva and plasma was notrelated to neutralizing antibody titers in serum or the animal’sclinical status (135). In contrast, adoptive transfer of sera fromFIV-infected or vaccinated cats was reported to effectivelyprotect cats against subsequent challenge with the homologousstrain of FIV (83). We cannot exclude, therefore, that in vivothese antibodies contribute to the limit of cell-to-cell spread,especially within nonlymphoid tissues.
Cell-Mediated Responses
Virus-specific, cell-mediated immune responses are pres-ently under close scrutiny in hosts infected with HIV and otherlentiviruses, especially with regard to their potential role invaccine-induced protection. By contrast, this topic has beenrather neglected in FIV research to date. There is only onepublished report dealing with FIV-specific cytotoxic T cells.Target cells consisted of autologous or allogeneic T-lympho-blastoid cells labelled with chromium-51 or indium-111. Freshor mitogen- and IL-2-stimulated PBMC from infected catsexerted modest levels of cytotoxicity, but their activity wasconsiderably increased following in vitro stimulation withirradiated, FIV-infected autologous lymphoblastoid cells. Theeffector cells lysed autologous but not allogeneic target cellsand were composed predominantly of CD81 T cells, indicatingthat the effect was mediated by major histocompatibility com-plex-restricted T cells. Cytotoxic T cells could be detected asearly as 7 to 9 weeks p.i. (213). Viral antigen-induced lympho-proliferation experiments have been reported for cats infected
with FIV (116) or immunized with experimental vaccines(257), but a response was seen only in the latter case.
PATHOLOGICAL EFFECTS
The pathological effects observed in FIV-infected cats maybe both a direct consequence of FIV infection and secondaryto the immunodeficiency state caused by the virus.
Clinical Manifestations
The development of accurate, commercially available sero-logic tests has made it possible to identify FIV-infected cats inthe field. The illnesses described in such animals are numerousand extremely varied and may be episodic but tend to becomepersistent with progression (172, 173, 176, 259). From theobservation of field and experimentally infected SPF cats, ithas become evident that the FIV disease can be divided intoseveral sequential steps. Ishida and Tomoda (96), on the basisof the type and severity of clinical signs, have proposed astaging of FIV-infected cats into five phases (Table 2) whichreflects the classification systems used for HIV-induced pathol-ogy. Such staging, which has been widely accepted, is outlinedbelow with the caveat that range and course of clinical signs areextremely variable in individual cats.Acute phase. In some cases primary infection is clinically
silent. More commonly, it manifests itself as a transient illness(1 to 4 weeks) with generalized lymphadenopathy, mild pyrex-ia, dullness, depression, anorexia, and a neutropenia that in afew cats may allow the development of life-threatening bacte-rial infections. Acute diarrhea, conjunctivitis, dermatitis, gin-givitis, and mild upper respiratory symptoms are also presentin more severely affected animals. A few deaths have beenrecorded, but the link to FIV was not clear. As these earlymanifestations often respond to supportive and antibiotictreatments, in veterinary practice their etiology frequentlyescapes recognition (10, 14, 31, 47, 68, 133, 235, 259).Asymptomatic phase. After the signs associated with pri-
mary FIV invasion have subsided, the infection usually remainsclinically inapparent for prolonged periods as occurs withmany other retroviruses. In experimentally infected cats, clear-cut permanent immunological and hematological alterationsdo not usually become detectable sooner than 1.5 to 2 yearsp.i., and clinical manifestations of immunodeficiency appearsubstantially later, even though infection is usually performedwith virus doses presumably much larger than those involved innatural transmission (1, 10, 135, 235). In fact, many SPF catsinfected over the past years in different laboratories withvarious isolates of the virus are still disease-free. This may bepartly because most such cats are maintained in pathogen-freefacilities, but epidemiological evidence indicates that in thefield also the infection tends to remain clinically silent for years(98, 172, 205). It is important to emphasize that clinicalnormality does not imply a virological latency, as shown by thefact that FIV can be isolated from the PBMC, plasma, and
TABLE 2. Stages in FIV infection
Stage Clinical manifestations Duration
Primary infection None or transient, usually mild constitutional symptoms Weeks to monthsAsymptomatic carrier None YearsPGLa Generalized lymphadenopathy and vague constitutional signs Months?ARC Lymphadenopathy, secondary but not opportunistic chronic infections, weight loss of ,20% Months to yearsFAIDS Severe secondary and opportunistic chronic infections, tumors, wasting Months
saliva with the same frequency as from sick cats (135). Thoughthe annual rate of progression from the asymptomatic to thesymptomatic state and eventually to feline AIDS (FAIDS) isstill unknown (95), it is clear that the evolution of infection isslow, even though from the acute phase a few cats may enterthe subsequent symptomatic phases without going through anasymptomatic period. It is noteworthy that the length of theincubation period does not seem to correlate with prognosis(133).Persistent generalized lymphadenopathy. Persistent gener-
alized lymphadenopathy is characterized by a long-lastinggeneralized enlargment of lymph nodes. Vague signs of dis-ease, including recurrent fevers, anorexia, weight loss, ornonspecific behavioral changes, are almost invariably present.Overt superinfections are usually absent (98, 172, 255).ARC. The acronym ARC, for AIDS-related complex, was
introduced in human medicine for indicating a number ofmanifestations prodromic to full-blown AIDS but not fulfillingthe definition of AIDS. The term has been progressivelyabandoned for humans but remains useful for FIV-infectedcats because the natural history of FIV-induced pathology isstill poorly understood and a general consensus on the defini-tion of FAIDS is still lacking. Cats in this category usuallypresent with chronic secondary infections of the oral cavity,upper respiratory tract, and other body sites but no opportu-nistic infections. The agents causing these infections are vi-ruses, including FeLV (35, 71, 90, 96, 202, 255), feline syncy-tium-forming virus (9, 255), feline calicivirus (41, 105, 225,246), feline infectious peritonitis virus (98), feline herpesvirus(193), feline papillomavirus (49), and poxvirus (24); bacteria,including Staphylococcus sp. (176), Pseudomonas sp. (98),Streptococcus canis (172), Yersinia pseudotuberculosis (77),nontuberculosis mycobacteria (98, 220), and other aerobic andanaerobic bacteria (71, 85, 90, 96, 126, 255); fungi, includingCandida albicans (98, 130), Cryptococcus neoformans (98, 130),and Microsporum canis (98, 130); protozoa, including Crypto-sporidiium sp. (152) and Hemobartonella felis (71, 85, 98); andparasites, including Toxoplasma gondii (40a, 98, 108, 167, 253),Dirofilaria immitis (261), Demodex canis (32, 176, 220), Noto-edres cati (98, 176), and Otodectes cynotis (214).Weight loss without marked emaciation, generalized lymph-
adenopathy, fever of undetermined origin, and hematologicabnormalities are common. Other symptoms reported includealopecia and pruritus. Neurologic, renal, neoplastic, and otherdisorders also may be present in a small proportion of cats.Most, if not all, cases diagnosed as ARC progress to FAIDSafter variable time intervals (85, 96, 172).FAIDS. FAIDS, similar in many respects to full-blown AIDS
of humans, has been observed in 5 to 10% of the clinically illFIV-infected cats brought to a veterinarian. They suffer fromsevere secondary infections listed above and, to a lesser extent,from neoplastic and neurologic disorders. The tumors de-scribed in FIV-infected cats include high- and low-gradelymphosarcomas, which are frequently extranodal (12, 29, 94,182, 202), fibrosarcomas (98), myeloproliferative diseases (94,98, 202), mast cell tumors (12, 202), squamous cell carcinomas(94, 173), miscellaneous adenomas and carcinomas (72, 85),oligodendroglioma (93), and meningioma (22). The neurologicabnormalities observed in FIV-infected cats are reviewed inreference 107 and include altered behavior and attitude,convulsions, nystagmus, ataxia, sterotypic motor behavior (re-petitive, compulsive roaming), intention tremors, dementia,paralysis (rare), abnormal slow motor and sensory nerveconduction velocities, electroencephalographic alterations,evoked potential abnormalities (prolonged interpeak laten-cies), magnetic resonance imaging abnormalities (cortical at-
rophy, ventricular enlargement, and white matter lesions),cerebrospinal fluid abnormalities (pleocytosis, increased IgGlevels, and anti-FIV antibody), and histopathological abnor-malities. Infections are often multiple, sustained by opportu-nistic agents, and resistant to treatment. Most animals presentwith a marked loss of body weight as well as severe anemia andleukopenia. The clinical picture worsens rapidly, and once thediagnosis is made, the mean survival time is usually less than 1year, in spite of supportive therapy (96, 172).In addition to the above stages, Pedersen (172) has proposed
a sixth group to include cats presenting with miscellaneousdisorders (neurological, neoplastic, ocular, and immunologi-cally mediated, etc.) in the absence of other manifestationsthat, if present, would allow their inclusion in the previousclassification. The usefulness of this all-inclusive additionalgroup is bound to diminish as the disease-inducing potential ofFIV is progressively better understood.
Hematological, Immunological, and Other Laboratory Signs
Although the staging of FIV infection is based principally onclinical criteria, the primary defect is the underlying immunedysfunction. This is clearly evident at the levels of bothhematological parameters and in vivo and in vitro measurableimmune functions.As in HIV infection, the hematological landmark of FIV
infection is a progressive depletion of the CD41 helper-inducer T-lymphocyte subset. In cats infected as young adults,the decline of circulating CD41 lymphocytes develops in twodistinct phases: an early decrease observed during primary infec-tion (47) is followed by a more gradual decline (Fig. 7). Thus, it isusually only by 1 or more years p.i. that clear-cut CD41 cellreductions can be unequivocally diagnosed in individual cats.Certain FIV isolates, however, appear to induce more markedand rapid perturbations of lymphocyte subsets than others (234).Eventually, in terminally ill animals, CD41 cell numbers maybe lower than 200/ml (normal range, 1,000 to 3,500/ml) (1, 10,68, 82, 159, 235). The threshold of CD41 T-cell drop that iscritical for evolution to FAIDS has yet to be established withcertainty, but counts below 200/ml are usually observed only indiseased cats.Infected cats also develop an inversion of the CD41/CD81
T-lymphocyte ratio (normal values, .2) due both to a CD41
cell decrease and, in some instances, to an expansion of CD81
suppressor-cytotoxic cells (1, 47, 68, 234, 235, 249). CD81 cellsalso display an upregulation of major histocompatibility com-plex class II antigen expression (163, 195, 249). On the otherhand, B-cell numbers remain essentially unchanged (1, 159,234).Other hematological abnormalities are also common (204).
In the primary phase a transient leukopenia, due to an absoluteneutropenia, is frequent (259). Cytopenias are usually absentin the asymptomatic phase but occur in a large proportion ofcats with ARC and FAIDS (203). These may be lymphopenias(,1,500 lymphocytes per ml), nonresponsive anemias (hema-tocrit, ,24%), neutropenias (,2,500 neutrophils per ml), and,less commonly, thrombocytopenias. Multiple concomitant cy-topenias are also common (16, 71, 72, 78, 85, 97, 98, 201, 204,205, 220, 255, 261). Leukocytosis due to lymphocytosis, mono-cytosis, and neutrophilia has also been reported in infected cats(85, 255).The serum immunoglobulin level (mainly IgG) is almost
invariably increased polyclonally in infected cats. This cannotbe explained simply by the presence of secondary infectionsbecause it occurs also in experimentally infected SPF cats notexposed to others pathogens (1, 85, 186). Elevated levels of
IgG were also found in saliva (186). The hypergammaglobu-linemia correlates with hyperactivation of the B-cell compart-ment in lymphoid organs as seen histologically (see below) andfrom enhanced levels of heterophile antibodies and autoanti-bodies (63). The levels of circulating immune complexes mayalso be enhanced (17). Other biochemical data are generallyunremarkable (85). Abnormalities observed in some cats, suchas hyperbilirubinemia (31), generally but not necessarily reflectsecondary or concomitant diseases involving specific organs(76). Increased blood urea nitrogen and serum creatinine (185,226) indicate a renal damage of unclear etiology (see below).Normal or enhanced antibody responses have been reported
(110, 224, 235), but usually FIV-infected cats mount reducedantibody responses to a variety of antigens and microorganisms(41, 172, 193, 224, 235). The defects become detectable severalmonths after infection, affect especially primary responses toT-cell-dependent antigens, and may be due partly to impair-ment of IgM-IgG switching. Symptomatic FIV-infected catsmay also exhibit elevated levels of spontaneous or provokedtissue necrosis factor alpha in plasma (110, 111) and of IL-2receptor alpha-positive PBMC (163).Functional abnormalities have also been observed by mon-
itoring immunocompetent cell performances in in vitro tests.These include reduced blastogenic and IL-2 responses ofPBMC to T and B mitogens (10, 20, 21, 75, 117, 208, 223, 235)and antigen-specific priming of naive T cells to soluble proteinantigens (21). Some such deficits are detectable soon afterinfection when circulating lymphocyte subsets are still normal(20, 95, 109, 223), while others tend to develop at later stages(163, 235). Interestingly, in vitro mitogen-induced IL-2 recep-tor alpha expression was depressed in both CD41 and CD81
cells, suggesting the existence of functional defects in the lattercell type also (163). Functional alterations have also beenobserved in macrophages, neutrophils, and natural killer cellfunctions of infected cats (74, 115).
Histopathological Changes
Histopathological changes in lymphoid tissues of FIV-in-fected cats are remarkably similar to those of HIV-infectedindividuals. Changes are not so specific as to permit thediagnosis of lentivirus-induced immunodeficiency; neverthe-less, they are useful for staging purposes. They vary greatly notonly depending on time from infection and other clinicalvariables but also between individual animals and within lymphnodes of the same animal (30, 172). There is, however, ageneral pattern that consists of an exuberant follicular hyper-plasia and follicular pleomorphism followed by progressivefollicular exhaustion and involution (Fig. 8).Cat lymph nodes resemble human ones in morphology and
cellular content (30). During the early stages of FIV infection,the lymph nodes tend to be hyperplastic and show the presenceof medium to large secondary follicles in the cortex. Mantlezones vary greatly in thickness, and germinal centers areincreased in number and size. Nodular or diffuse expansions ofthe paracortical T-cell zone and hypercellularity of the sinusescontaining lymphoid and histiocytic cells have also been ob-served. In the asymptomatic phase, the follicular hyperplasiamay be temporarily reduced, while in the ARC phase it mayagain be noted, concurrent with involution. The latter changesinclude obscure follicular margins, follicular atrophy and seg-mentation, and hyalinization of the germinal centers. The finalstages of the disease process are characterized by markedinvolution and depletion of the lymph nodes. The corticalfollicles become cell depleted or not discernible at all and maycontain aggregates of hyaline eosinophilic material; the mantlezones are often indistinct, the paracortical regions are reducedor absent, and the sinuses are collapsed and empty (14, 25, 30,133, 194). It should be noted, however, that in a blind study themain feature of FIV-positive lymph nodes, compared withFIV-negative ones, was a higher degree of plasmacytosis (194).
FIG. 7. Mean absolute CD41 T-lymphocyte counts, mean absolute CD81 T-lymphocyte counts, and mean CD41/CD81 T-lymphocyte ratios of cats infected withFIV at birth, as young adults, or in old age. Datum points with a statistically significant difference between infected (E) and noninfected cats (F) are designated withan asterisk (reproduced from reference 68 with permission of the publisher).
Plasma cells were seen to increase in number both within andaround germinal centers with progression of infection (30). Inone report of lymphocyte subsets in lymph nodes of FIV-positive cats, a slight increase of CD81 lymphocytes in theparacortical regions was seen, with no CD81 cells withinfollicles (194). In another study, increased numbers of CD81
cells were found in germinal centers throughout the year ofobservation p.i., while CD41 cell levels initially showed anincrease but returned to normal levels within months (30).These aspects should be further investigated.Splenic changes appear to parallel the lesions observed in
the lymph nodes, while the thymus may show atrophy (14). Thebone marrow of a high proportion of symptomatic cats re-vealed hypercellularity due to hyperplasia of individual celllines, maturation abnormalities, lymphoid infiltrates, plasmo-cytosis, excess eosinophils, and areas of necrosis (14, 131, 204).In nonlymphoid tissues histopathological lesions are mainly
evident in the advanced stages of infection. In the brain,however, moderate gliosis and glial nodules, sometimes asso-ciated with perivascular infiltrates, and white matter pallorwere observed as early as 1 or 2 months p.i. depending onwhether the virus had been injected intracerebrally or intrave-nously. Such lesions remained unchanged for 1 year p.i., but innaturally infected animals, presumably infected for longerperiods, they were more severe and often accompanied bysigns of meningitis (46, 93, 106). These lesions are similar tothe neuropathological findings in people with AIDS (4),though differences have been noted, including fewer or nomultinucleate giant cells and a preferential distribution to thesubcortical structures in FIV-positive brains. Demyelinizationof the spinal cord and vacuolar myelopathy of the dorsal andventral nerve roots with focal mononuclear infiltrates have alsobeen described (248).Kidney abnormalities are frequent and varied. Mesangial
widening, with segmental to diffuse glomerulosclerosis, Ig(mainly IgM) and C3 deposits in the mesangium, and severe
tubulo-interstitial lesions consisting of diffuse interstitial infil-tration by lymphocytes and plasma cells, as well as fibrosis andtubular degenerative changes, have been described in naturallyinfected animals (185), and glomerular mesangial cell prolif-eration has been seen in experimentally infected cats (17, 133).Recent evidence by Poli et al. (184) indicates that kidney andother organs of FIV-infected cats may also contain amyloiddeposits.Oral lesions are common and usually consist of mild or
proliferative gingivitis and periodontitis with or without in-volvement of extragingival tissues with lymphocyte and plasmacell infiltration. Prominent signs of transmural typhlitis havebeen described in SPF cats experimentally infected 2 or moremonths earlier (14). Cats with chronic diarrhea often showsevere and diffuse villous atrophy, particularly evident in thelower intestine and focal areas of mural necrosis, with fibrosisand hyperplasia of the associated lymphoid tissue (25, 31, 259).Degenerative cardiomyopathy and interstitial pneumonitishave also been seen in recently infected cats (14, 77), whilelesions of the liver (cholangiohepatitis, periportal fibrosis, andbiliary stasis) and other organs may be present in sick animals(31, 44).
USEFULNESS OF FIV AS A MODEL FOR AIDS
It is generally accepted that the areas of AIDS research thatcan profit most from the study of animal models are (i)illuminating pathogenesis, (ii) development and improvementof antiviral therapies, and (iii) development and testing ofprotective vaccines. The status in these fields of FIV researchis briefly examined below.
Getting a Better Understanding of Pathogenesis
Despite recent advances, our understanding of the HIVdevelopmental cycle in infected individuals and of mechanisms
FIG. 8. Histopathological changes in lymph nodes of FIV-infected cats (stained with hematoxylin and eosin; magnification, 325). (A) Exuberant follicularhyperplasia with prominent and irregular follicular expansion, as observed in early stages of infection. (B) Follicular involution with sparsely cellular germinal centersand poorly delineated mantle zone, as observed in advanced stages of infection.
of HIV-induced diseases remains obscure, leading to a consid-erable uncertainty in the design of rational therapeutic ap-proaches (39, 57, 113, 247). Studies aimed at understandinghow FIV produces pathological effects are still limited, so thatit is not certain which paths are most useful to investigate andwhich will guide us to learn what is most urgent about lentiviralinfections of humans.One major issue for study is the role of the direct detrimen-
tal action of the virus on immunocompetent cells versusindirect cytotoxic mechanisms in the events leading to CD41
T-cell depletion and other immunological dysfunctions. Invitro, FIV infection of T lymphocytes is followed by cell lysis ina matter of a few days (discussed above). As reported byBishop et al. (19), ex vivo, PBMC from cats infected for 2 to 5years undergo morphological changes and genomic DNAfragmentation characteristic of apoptosis in the absence of anyadded stimulation. Whether these phenomena also occur toany significant extent in vivo is not known. Although thenumber of provirus-carrying PBMC is relatively high through-out infection (136, 138), the long incubation period and theprolonged course of FIV disease suggest that in the intactanimal virus expression must be under some vigorous form ofrestraint. Indeed, in tissues infected long term, relatively fewcells express FIV molecules that are detectable by currentlyavailable immunological and molecular techniques (40, 122).Thus, unless the picture is dramatically changed by the devel-opment of more sensitive techniques, serious attempts shouldbe made to understand the factors that keep FIV expressioncontained so efficiently in the intact animal and unable to domuch damage for such prolonged periods of time. Infection isaccompanied by a remarkable activation of B cells and theearly histological changes are dominated by this phenomenon.Inappropriate activation of the immune system occurs also inHIV infection and has been suggested to represent a funda-mental abnormality, with immunosuppression being only sec-ondary to it (39, 57). The mechanisms of B-cell stimulation byFIV and the antigenic specificity of activated cells, if any, haveyet to be determined (63). It may be worth mentioning that arecombinant peptide incorporating the V3 and V4 regions ofFIV SU protein was found to be mitogenic in vitro for naivefeline but not human lymphocytes (13).Similar considerations apply to the hematopoietic alter-
ations that may be so prominent during FIV infection. Theonly bone marrow cells that have been shown to harbor thevirus are megakaryocytes and macrophages (15, 122). Inaddition, bone marrow cell progenitors of infected cats showeda reduced responsiveness in vitro to hemopoietic growthfactors, which seemed to reflect the presence of inhibitorysubstances in serum, but the part these alterations play in thehost is not known (118, 201).Regarding the numerous other clinical and pathological
manifestations associated with FIV, a major problem is todiscriminate what is the precise contribution of recognized orunrecognized secondary infections. A direct FIV etiologyseems most likely in those pathological lesions and clinicalmanifestations which are observed at high frequencies ininfected cats. Neurological symptoms dominate the picture inabout 5% of otherwise asymptomatic cats and accompanyother manifestations in a similar proportion of clinically illanimals (78, 205, 220, 243, 255). While postmortem studieshave shown that in naturally infected cats CNS involvementoccurs much more frequently than clinically appreciable neu-rological deficits (46), kittens infected at 8 months of ageshowed neuropathophysiological changes analogous to HIVpediatric encephalopathy (181). As already discussed, FIV canbe isolated from the CNS of infected cats in which very high
provirus loads can be reached (180); the virus replicatesprimarily in astrocytes and microglia cells (45). Thus, it seemslikely that the neuropathological changes observed in infectedcats are mainly, if not solely, due to FIV replication in theCNS, but damage to neuronal functions is mainly indirect.Neurological abnormalities, however, may also stem fromsuperimposed opportunistic infections (81).Signs of renal disease have been observed in a high propor-
tion of FIV-infected cats (16, 97, 185, 220, 226). For example,a recent examination of 326 sick cats living in Australiademonstrated a significant association between FIV infectionand hyperazotemia and palpably small kidneys (226). The viruswas easily recovered from renal tissue by isolation and PCR,and occasional tubular epithelial cells as well as scatteredinterstitial inflammatory cells and glomerular cells stainedpositive for FIV CA antigen (14, 184). Thus, it also seemslikely that the renal lesions are at least partly a product of localFIV replication in the kidney. However, renal lesions also arefrequently observed at postmortem examination in nonin-fected aged cats (172), so it is also possible that FIV infectionsimply facilitates the damaging factors responsible for suchlesions.In the field, FIV infection is also associated with a wide
range of malignant neoplasms. Tumors are primarily of lym-phoid and myeloid origin, but miscellaneous sarcomas andcarcinomas also occur, some of which are rare in FIV-negativecats (172). In different reports, 1 to 28% of FIV-infected catswere found to bear tumors not attributable to FeLV (58, 62,202, 255, 259). In FIV-positive, FeLV-negative cats the likeli-hood of developing leukemia or lymphoma is considerablygreater than in noninfected cats, though manifold lower thanin cats singly infected with FeLV (202). Moreover, lympho-and myeloproliferative disorders have been observed in exper-imentally infected SPF cats free of known tumorigenic viruses(29, 182). As in AIDS patients (3), the great majority oflymphoid tumors are of B-cell origin. However, as for HIV,studies so far have failed to incriminate FIV in the moleculargenesis of such tumors and alternative mechanisms (reducedimmunosurveillance, activation of oncogenic viruses, and ex-cess activation of B cells, etc.) have been suggested (29, 182).Several attempts have been performed or are under way to
evaluate how putative constitutional and exogenous cofactorsinfluence the clinical course of FIV infection. Great (andpossibly unjustified) emphasis has been placed on cofactors inthe pathogenesis of human AIDS, and due to the prolongedclinical latency before full-blown disease, FIV is an idealsystem for these types of studies. A recent investigationexamined the effect of age at the time of infection on FIVdisease progression over a 10-month period. Cats infected asnewborns developed a more profound lymphadenopathy andneutropenia than the other age groups examined. In contrast,lymphocyte subset changes and overall illnesses were moresevere in aged cats. Neonates developed ELISA antibodyresponses to FIV comparable to young adults, while aged catsresponded more slowly (Fig. 7). The conclusion was that agingmay be an important determinant in FIV pathogenesis (68).Age at the time of infection also appeared to influence thevirus burdens found in PBMC (138).Upon superinfection with FIV, asymptomatic cats with
preexistent FeLV infection manifested a more accelerated andexacerbated FIV disease and showed higher FIV loads thandid naive cats under both natural and experimental conditions(71, 147, 175). This interrelationship is similar to what has beendescribed for HIV and human T-cell lymphotropic virus type 1(113). Interestingly, the synergy between FIV and FeLV isbidirectional: doubly infected cats developed FeLV-induced
tumors more frequently than did cats infected with either virusalone (202). In addition, FIV-induced histopathological lesionswere more evident in cats coinfected with FeLV or felineinfectious peritonitis virus than in cats singly infected with FIV(15, 93, 98). A similar, though less pronounced effect was seenin cats repeatedly immunized with unrelated antigens (112). Asynergistic suppression of mitogen responsiveness was ob-served in cats concurrently infected with FIV and T. gondii(116). The clinical significance of findings showing that felineviruses, such as the feline herpesvirus 1 which also infects Tlymphocytes, can transactivate FIV LTR is currently unknown(101, 102).Contrasting with the above findings, no increase in FIV-
induced pathology was seen in cats intentionally exposedsequentially to various pathogens and immunogens over aperiod of 3 years. As some of the parameters studied wereactually better in these animals than in nonmanipulated FIV-infected cats, the suggestion was raised that periodic immunestimulation might in fact ameliorate some of the deleteriouseffects of FIV-induced immunosuppression (192). No worsen-ing of FIV disease was noted in cats superinfected with felinesyncytium-forming virus (262). Further studies are warrantedto better delineate the role of these and other putativecofactors in FIV infection.
Developing and Improving Therapies
Currently available anti-HIV therapies are only partiallyeffective in controlling virus replication, and their action isoften transient because of the frequent emergence of drug-resistant viral mutants (99). In addition, the recent realizationthat HIV replication is more active through the entire courseof infection than formerly believed has raised the dilemma ofwhether antiviral treatment should be initiated early afterinfection or in more advanced HIV disease. Deciding is adelicate process because the advantages of starting therapyearly must be weighed against the adverse side effects of theavailable therapies and the enhanced likelihood of favoring anearly emergence of drug-resistant strains (57).The use of animal models is an essential step in the
development of antiviral drugs and combination therapies.With the caveat that even the best animal model is not 100%predictive of human responses, studies in the FIV system cancontribute information about the pharmacokinetics and phar-macodynamic properties of candidate anti-HIV compoundsand identify potential toxicities, as well as help establish theirantiviral potency and optimize the design of clinical protocols.FIV-induced full-blown disease is unsuitable as a criterion forevaluating efficacy of treatments in experimentally infectedanimals because of the prolonged interval between exposure tothe virus and clinical manifestations, although clinical criteriahave been advantageously used in naturally infected sick cats.However, as recently pointed out by Barlough et al. (11),infected cats need not be monitored for prolonged periods,since changes in clinical and hematological abnormalitiesevident within the first months of infection are sufficient tocompare the effects of treatments and other variables. Addi-tional parameters to monitor in recently infected animals arecirculating CD41 lymphocyte counts, which reproducibly de-cline before severe clinical deterioration, and viral burdensdetermined by culture or molecular techniques.As shown in Table 3, several potential anti-AIDS substances
and others already in use have been screened for the ability toinhibit FIV replication and FIV-induced cellular damage invitro. These include chain termination agents and a number ofsubstances whose mechanisms of action are less well under-
stood. The concentrations required to exert an anti-FIV effecthave generally been in the same range as those needed toinhibit HIV. The patterns of cross-resistance to nucleosideanalogs of 39-azido-29,39-dideoxythymidine (AZT)-resistantmutants of FIV were similar to those of AZT-resistant mutantsof HIV (190). These findings are in line with what is knownabout the susceptibility of FIV RT to inhibitors (156, 158) anddemonstrate that preclinical in vivo testing of new drugs andnew therapeutic protocols are useful approaches in the FIVmodel. This conclusion is confirmed by the published studiesthat have tested the prophylactic and therapeutic actions ofdrugs in FIV-infected animals (Table 4). In a recent investiga-tion by Remington et al. (191), two AZT-resistant mutants ofFIV reverted rapidly to susceptibility when passaged in theabsence of the drug. In the revertants, the single amino acidchange responsible for AZT resistance was maintained, noother base changes were noted in the RT coding region, andpurified RT remained AZT resistant. These findings suggestthe involvement of another genomic region, located outsidethe RT region, in AZT resistance of FIV. No studies oninhibitors of other viral enzymes have yet been reported.However, the recent chemical synthesis of the PR moleculeand its renaturation to an enzymatically active form (55)represent an essential step in the design of PR inhibitors withpossible therapeutic applications.
Criteria for Vaccine Development
The development of safe vaccines capable of conferringlong-lived and broad-spectrum protection against HIV is prov-ing extremely arduous. After nearly 10 years of research, keyquestions such as the types of immune responses it would bepreferable to elicit (TH1 versus TH2 or humoral versus cellmediated), the role exerted by antiviral antibody (protectiveversus infection enhancing), the set of antigens a vaccineshould incorporate (viral envelope versus internal componentsor both), the kind of the vaccine (technologically modernversus traditional), and even the target to pursue (protectionfrom infection versus protection from disease) remain essen-tially unanswered. More explicitly, it is still uncertain whetheran effective immune prophylaxis against HIV can indeed beachieved by vaccination (34, 80, 199). The main purpose ofusing animal models in this area is to provide sound conceptualgrounds on which to design candidate anti-HIV vaccines. Inthis respect, a system such as FIV in which a virus that closelyresembles HIV is studied in its natural host species should bemost informative. The fact that an efficacious anti-FIV vaccinewould be advantageous in veterinary medicine adds value tothe system.As clearly emerges from the previous sections, in many areas
our present understanding of FIV lags behind knowledgeaccumulated on HIV. This is inevitable, given the shorter timeFIV has been known and the lower number of laboratoriesinvolved in FIV research. In contrast, the number of vaccina-tion experiments performed with FIV, using both traditionaland subunit vaccine formulations, is relatively high. Unfortu-nately, most vaccines have failed to induce significant levels ofprotection, thus confirming the difficulty of developing effec-tive antilentiviral vaccines (171). Failures have involved vac-cines composed of inactivated whole virus, fixed whole infectedcells, recombinant CA and SU proteins, and a syntheticpeptide corresponding to the V3 region of the SU moleculeknown to contain a neutralizing antibody-eliciting epitope,delivered in a variety of adjuvants (Table 5). In some suchattempts, FIV infection was actually facilitated in vaccinatedanimals. Thus, cats vaccinated with purified FIV incorporated
into immunostimulating complexes (ISCOM), recombinantp24 in ISCOM, or glutaraldehyde-fixed infected PBMC mixedwith the adjuvant quil A became infected more frequently andearlier than unvaccinated untreated cats. As no enhancingantibody was observed in the vaccinated cats, the effect wastentatively attributed to the activation of the immune systemassociated with repeated immunization (88). A potentiation ofchallenge infection was also noted in cats vaccinated with thesynthetic V3 peptide mentioned above (17). Enhancement ofinfection has also been seen in certain SIV vaccine experiments(219).Thus, the FIV vaccine scenario would be completely bleak if
not for the results reported by Yamamoto and her group. In afirst study (257), these workers developed two vaccines, oneconsisting of paraformaldehyde-fixed infected cells and theother consisting of paraformaldehyde-inactivated whole virus,which were administered subcutaneously to SPF cats four to sixtimes over 2 to 5 months. The adjuvants used were muramyl
dipeptide and a combination of complete and incompleteFreund’s adjuvant, respectively. Both vaccines were safe andnontoxic and induced a strong FIV-specific immune response,as shown by ELISA and neutralizing antibodies, as well asproliferation and IL-2 production by PBMC stimulated in vitrowith inactivated whole FIV. Vaccinated and control cats thathad received either placebo or noninfected cells were chal-lenged intraperitoneally with the homologous virus 2 weeksafter the last immunization. As judged by virus isolation andPCR, 13 of 15 vaccinated cats were protected, while allcontrols were infected. These findings were recently confirmedand expanded to show that vaccinated cats were protected alsoagainst a heterologous strain of FIV, differing by 11 and 4% inthe amino acid sequence of the SU and TM proteins, respec-tively, from the vaccine virus (256), and that immune serataken from vaccinated cats were also able to protect againstFIV infection upon passive transfer into naive animals (83).Most recently, in a preliminary report, Hosie et al. (89), using
TABLE 3. Established and potential anti-AIDS compounds tested for anti-FIV activity in vitro
Compound Parameter(s) examined Result Reference(s)
HP 0.35a FIV replication in PBMC Inhibition 73
AZT FIV replication in thymocytes Inhibition 48Cell killing Inhibition
FIV replication in lymphoblastoid cells Inhibition 222Cell killing InhibitionFusogenic activityb No effect
FIV replication in lymphoblastoid cells Inhibition 137
PMEAc FIV replication in thymocytes Inhibition 48FIV replication in CrFK cells Inhibition 37
ddI FIV replication in lymphoblastoid cells Inhibition 222Cell killing InhibitionFusogenic activity No effect
ddA FIV replication in CrFK cells Inhibition 37
Pradimicina A FIV replication in lymphoblastoid cells Inhibition 222Cell killing InhibitionFusogenic activity Inhibition
Heparin FIV replication in lymphoblastoid cells Inhibition 222Cell killing InhibitionFusogenic activity No effect
4-Amino-3,6-disulfonato-1,8-naphthalimide FIV replication in lymphoid cells Inhibition 196
ddCTP, encapsulated in cat erythrocytes FIV replication in macrophages Inhibition 127, 128
Derivatives of 4-amino-3,6-disulfonate-1 FIV replication in CrFK cells Inhibition 196
8-Naphthalimide Cell killing Inhibition 196
a Cephalosporin degradation product.b Determined by coculturing FIV-infected cells with human T-cell lymphotropic virus type 1-positive human MT-2 cells.c PMEA, 9-(2-phosphonylmethoxyethyl)adenine.
a similar fixed infected-cell vaccine, confirmed the protectiveeffect against homologous virus, but their vaccine was unableto protect against heterologous virus challenge.The reasons for the discrepancies observed in different
vaccination experiments reported above are not clear. Eventhough data indicate that protection can be achieved withoutthe apparent involvement of responses to cellular antigens(92), in the SIV-macaque model the incorrect use of xenoge-neic cells as substrates for producing both the vaccine and thechallenge virus has led to misleading results, that is, toprotection mediated by immune responses to cellular antigensinstead of, or in addition to, immune responses elicited by viral
antigens (36). The incorporation of host cell proteins intolentivirus envelopes is well documented (2, 168). In thesuccessful FIV studies summarized above, the virus was prop-agated in feline cells, thus ruling out a role for xenogeneicantigens in protection. In addition, immunization with the cellsused to propagate the virus alone did not protect, and nocorrelation was noted between protection and the antibodyresponse generated against feline cells. Thus, these resultsseem to support the concept that protective immunity can beachieved with antilentivirus vaccines. However, a note ofcaution is necessary because the successful FIV vaccinationexperiments described above have in common the use of FIV
TABLE 4. Established and potential anti-AIDS compounds tested for anti-FIV activity in vivo
Compound Parameter(s) examined Result Reference(s)
AZT Clinical symptoms Improvement 79CD41/CD81 ratio Improvement
AZT FIV burden in plasma Transient reduction 137FIV burden in PBMC No effect
AZT, started before infection Circulating CD41 cells Slowed decline 107FIV burden in PBMC Reduced
PMEAa Clinical symptoms Improvement 79CD41/CD81 ratio Improvement
PMEA, started before infection FIV burden in PBMC ReductionClinical symptoms Improvement 48
PMEA, started before infection FIV burden in PBMC Reduction 179
Cyclosporine FIV burden in plasma Transient reduction, followed by increase 137
FIV burden in PBMC No effect
ddCTP, encapsulated in cat erythrocytes FIV burden in macrophages Reduction 127, 128FIV antigen-positive PBMC Reduction
a PMEA, 9-(2-phosphonylmethoxyethyl)adenine.
TABLE 5. Vaccination experiments performed with FIV
Vaccine ChallengeaProtection
Reference(s)bHomologous Heterologous
Paraformaldehyde-fixed infected cells Parenteral Yes NDc 257Paraformaldehyde-inactivated cell-free pelleted whole virus Parenteral Yes ND 257Detergent-disrupted, gradient-purified, envelope-depleted virus Parenteral No ND 88Recombinant p24 Parenteral No ND 88Glutaraldehyde- or beta-propriolactone-fixed infected cells Parenteral No ND 86, 88Paraformaldehyde-fixed infected cells Parenteral Yes Yes 256Paraformaldehyde-inactivated cell-free pelleted whole virus Parenteral Yes Yes 256Paraformaldehyde-fixed infected cells Parenteral Yes Yes 76Paraformaldehyde-fixed infected cells Mucosal No No 76Paraformaldehyde-fixed infected cells Parenteral Yes No 89Recombinant env gene product (Escherichia coli) Parenteral ND No 125Recombinant env gene product (baculovirus) Parenteral No No 125Paraformaldehyde-fixed infected cells or whole virus Mucosal Partial ND 188Paraformaldehyde-fixed infected cells Parenteral No ND 242Synthetic peptide of env V3 region Parenteral No ND 121
a Challenge was usually performed with 10 to 20 50% cat infectious doses.b Because this is a rapidly growing field, extensive reference is made to recently presented abstracts.c ND, not determined.
(Petaluma strain) propagated in vitro for prolonged periods asa source of immunizing antigen and challenge virus. Thepossibility that prolonged growth in the absence of immuneselective forces has reduced FIV virulence or antigenic poly-mophism (34, 247) should be considered. Also, a role forallogeneic cell antigens cannot be entirely excluded as thetiters of neutralizing antibody were not predictive in distin-guishing protected from nonprotected animals (256, 257). Thepossibility of successfully vaccinating against FIV should there-fore be tested by challenging vaccinated cats with in vivo-propagated strains of FIV identical to those circulating innature. If confirmed, the efficacy of vaccine-conferred protec-tive immunity against cell-associated virus, its duration (anti-viral antibody levels declined rapidly in the protected cats[257]), and its correlates should be assessed. Finally, theeffectiveness of the candidate vaccine should be evaluated inthe field. Because of the occurrence of natural disease, noother lentiviral model can lend itself to the latter purposebetter than FIV.
CONCLUDING REMARKS
With the increasing use of genetically and immunologicallymanipulated animals, the list of possible animal models ofAIDS is expanding (103). While it is likely that each of themodels proposed will provide useful information in specificareas of research, those that in recent years have rankedhighest are undoubtedly SIV (43) and, at a certain distancebehind, FIV (67). As is to be expected for any animal model,both viruses show some dissimilarities from HIV, but on thewhole, their biology, infection cycles, and diseases closelymimic those of HIV (Table 6). The features that have renderedSIV more appealing than FIV as a model system are its closergenetic relatedness to HIV, the fact that the host presumablyreacts in a manner more similar to humans than nonprimateanimals, and the relatively short incubation period before theonset of overt disease. However, the FIV-cat system, being thesmallest natural model of lentivirus infection, has a number oflogistical and practical assets that should not be overlooked(Table 6), and as discussed above, has already proven to be apowerful tool for deciphering the biology, pathogenesis, andsusceptibility to drugs and vaccine-induced immunity of immu-nodeficiency-inducing lentiviruses. The only real logisticaldrawbacks of studies with FIV are a shortage of immunologicalreagents and the relatively poor knowledge of the immunesystem of cats (114). These two shortcomings are rapidlychanging (252) and could be overcome by appropriate foster-ing of FIV research. Compared with SIV, which is carried as anonapparent infection in its natural host species, FIV repre-sents a significant affliction for its natural host. Thus, experi-ments aimed at a better understanding of the virus and thedevelopment of effective prophylactic and therapeutic agentswill affect only a small number of animals compared with thecat populations that would benefit from the findings. It ishoped that these considerations will help provide furtherimpetus to FIV research.
ACKNOWLEDGMENTS
We are grateful to the many colleagues of the FIV researchcommunity who have helped us in establishing the FIV model in ourlaboratories and have made available preprints of their articles. Theirnames are cited many times throughout the paper and need not bementioned again here.This work was supported by grants from the Ministero della
Sanita-Istituto Superiore di Sanita ‘‘Progetto Allestimento ModelliAnimali per l’AIDS’’; the Ministero Universita e Ricerca Scientifica,
TABLE 6. Comparison of the FIV-cat and SIV-macaque systems asmodels for AIDS studies
Factor FIV SIV
Logistical and operational aspectsRisk of contagion for operators No YesNatural virus-host system Yes NoAvailability of naturally sick hosts Abundant NoSupply of animal hosts Abundant ScarceCost of animal hosts Moderate HighHousing and handling of animal hosts Easy ComplexUnderstanding of host’s immune system Poor PoorAvailability of reagents Scarce Good
Similarity to HIV in:Viral biologyGeneral understanding Improving GoodAvailability of infectious molecularclones
Yes Yes
Virion morphology and biochemistry Yes YesPercent similarity in nucleotidesequence
Low About 50%
Genome organization Partial YesGenetic variation among isolates Yes YesGrouping in envelope sequencesubtypes
Yes Yes
Major cellular receptor CD9?a CD4
Replicative cycle ? YesFusogenic activity Yes YesPresence of linear neutralizing domainin V3
Yes No
Sensitivity to RT inhibitors Yes Yes
In vivo infectionGeneral understanding Improving GoodSexual transmission ? ?Perinatal transmission Rare RareExperimental mucosal transmission Yes YesReplication in CD41 T cells Yes YesReplication in other lymphocytes ? NoReplication in macrophages Yes YesReplication in dendritic cells Yes YesEarly invasion of CNS Yes YesAntigenemia Yes YesAntiviral immune responses Vigorous IrregularAntigenic drift Likely YesIndefinite persistence Yes YesIncreased virus load with progression Likely YesExistence of virulence variants Yes Yes
Induced diseaseGeneral understanding Good GoodPrimary disease Irregular IrregularIncubation before full-blownpathology
Years Months
Variable clinical progression Yes YesDiseases heralded by CD41 T-cell loss Yes YesSecondary and opportunisticinfections
Yes Yes
Kaposi’s sarcoma No NoB-cell lymphomas and other tumors Yes YesNeurological manifestations Yes YesWasting Yes YesExacerbation by cofactors Yes ?
Rome, Italy; and the European Commission Concerted Action onFeline AIDS Vaccination.
REFERENCES1. Ackley, C. D., J. K. Yamamoto, N. Levy, N. C. Pedersen, and M. D. Cooper.1990. Immunologic abnormalities in pathogen-free cats experimentallyinfected with feline immunodeficiency virus. J. Virol. 64:5652–5655.
2. Arthur, L. O., J. W. Bess, Jr., R. C. Sowder II, R. E. Benveniste, D. L. Mann,J.-C. Chermann, and L. E. Henderson. 1992. Cellular proteins bound toimmunodeficiency viruses: implications for pathogenesis and vaccines.Science 258:1935–1938.
3. Astrin, S. M., E. Schattner, J. Laurence, R. I. Lebman, and C. Rodriquez-Alfageme. 1992. Does HIV infection of B lymphocytes initiate AIDSlymphoma? Detection by PCR of viral sequences in lymphoma tissue. Curr.Top. Microbiol. Immunol. 182:399–407.
4. Atwood, W. J., J. R. Berger, R. Kaderman, C. S. Tornatore, and E. O.Major. 1993. Human immunodeficiency virus type 1 infection of the brain.Clin. Microbiol. Rev. 6:339–366.
5. Avrameas, A., J.-G. Guillet, L. Chouchane, A. Moraillon, P. Sonigo, andA. D. Strosberg. 1992. Localization of three epitopes of the env protein offeline immunodeficiency virus. Mol. Immunol. 29:565–572.
6. Avrameas, A., A. D. Strosberg, A. Moraillon, P. Sonigo, and G. Pancino.1993. Serological diagnosis of feline immunodeficiency virus infection basedon synthetic peptides from Env glycoproteins. Res. Virol. 144:209–218.
7. Baldinotti, F., D. Matteucci, P. Mazzetti, C. Giannelli, P. Bandecchi, F.Tozzini, and M. Bendinelli. 1994. Serum neutralization of feline immuno-deficiency virus is markedly dependent on passage history of the virus andhost system. J. Virol. 68:4572–4579.
8. Ball, J. M., K. E. Rushlow, C. J. Issel, and R. C. Montelaro. 1992. Detailedmapping of the antigenicity of the surface unit glycoprotein of equineinfectious anemia virus by using synthetic peptide strategies. J. Virol.66:732–742.
9. Bandecchi, P., D. Matteucci, F. Baldinotti, G. Guidi, F. Abramo, F. Tozzini,and M. Bendinelli. 1992. Prevalence of feline immunodeficiency virus andother retroviral infections in sick cats in Italy. Vet. Immunol. Immuno-pathol. 31:337–345.
10. Barlough, J. E., C. D. Ackley, J. W. George, N. Levy, R. Acevedo, P. F.Moore, B. A. Rideout, M. D. Cooper, and N. C. Pedersen. 1991. Acquiredimmune disfunction in cats with experimentally induced feline immunode-ficiency virus infection: comparison of short-term and long-term infections.J. Acquired Immune Defic. Syndr. 4:219–227.
11. Barlough, J. E., T. W. North, C. L. Oxford, K. M. Remington, S. Dandekar,M. N. Ellis, and N. C. Pedersen. 1993. Feline immunodeficiency virusinfection of cats as a model to test the effect of certain selection pressureson the infectivity and virulence of resultant lentivirus variants. AntiviralRes. 22:259–272.
12. Barr, M. C., M. T. Butt, K. L. Anderson, D.-S. Lin, T. F. Kelleher, and F. W.Scott. 1993. Spinal lymphosarcoma and disseminated mastocytoma associ-ated with feline immunodeficiency virus infection in a cat. J. Am. Vet. Med.Assoc. 202:1978–1980.
13. Beatty, J. A., G. Reid, M. A. Rigby, J. C. Neil, O. Jarrett, and M. J.Browning. 1992. A recombinant feline immunodeficiency virus envelopefusion protein stimulates peripheral blood lymphocytes from naive cats toproliferate in vitro. Vet. Immunol. Immunopathol. 35:143–153.
14. Beebe, A. M., N. Dua, T. G. Faith, P. F. Moore, N. C. Pedersen, and S.Dandekar. 1994. Primary stage of feline immunodeficiency virus infection:viral dissemination and cellular targets. J. Virol. 68:3080–3091.
15. Beebe, A. M., T. G. Gluckstern, J. George, N. C. Pedersen, and S.Dandekar. 1992. Detection of feline immunodeficiency virus infection inbone marrow of cats. Vet. Immunol. Immunopathol. 35:37–49.
16. Belford, C. J., R. I. Miller, G. Mitchell, R. S. Rahaley, and V. H. Menrath.1989. Evidence of feline immunodeficiency virus in Queensland cats:preliminary observations. Aust. Vet. Pract. 19:4–6.
17. Bendinelli, M., M. Pistello, S. Lombardi, A. Poli, C. Garzelli, D. Matteucci,and F. Tozzini. Unpublished data.
18. Bendinelli, M., M. Pistello, D. Matteucci, S. Lombardi, F. Baldinotti, P.Bandecchi, R. Ghilarducci, L. Ceccherini-Nelli, C. Garzelli, A. Poli, F.Esposito, G. Malvaldi, and F. Tozzini. 1993. Small animal models of AIDSand the feline immunodeficiency virus. Adv. Exp. Biol. Med. 335:189–202.
19. Bishop, S. A., T. J. Gruffydd-Jones, D. A. Harbour, and C. R. Stokes. 1993.Programmed cell death (apoptosis) as a mechanism of cell death inperipheral blood mononuclear cells from cats infected with feline immu-nodeficiency virus. Clin. Exp. Immunol. 93:65–71.
20. Bishop, S., N. A. Williams, T. J. Gruffydd-Jones, D. A. Harbour, and C. R.Stokes. 1992. An early defect in primary and secondary T cell responses inasymptomatic cats during acute feline immunodeficiency virus (FIV) infec-tion. Clin. Exp. Immunol. 90:491–496.
21. Bishop, S. A., N. A. Williams, T. J. Gruffydd-Jones, D. A. Harbour, andC. R. Stokes. 1992. Impaired T-cell priming and proliferation in catsinfected with feline immunodeficiency virus. AIDS 6:287–293.
22. Bo, S., P. Buracco, R. Guglielmino, A. Ratto, A. Bonoli, F. Valenza, andM. G. Canese. 1993. Personal communication.
23. Bosch, M. L., K. H. J. Siebelink, and A. D. M. E. Osterhaus. 1993. Genomeorganization and genetic variation of feline immunodeficiency virus. AIDSRes. Rev. 3:21–39.
24. Brown, A., M. Bennett, and C. J. Gaskell. 1989. Fatal poxvirus infection inassociation with FIV infection. Vet. Rec. 124:19–20.
25. Brown, P. J., C. D. Hopper, and D. A. Harbour. 1991. Pathological featuresof lymphoid tissues in cats with natural feline immunodeficiency virusinfection. J. Comp. Pathol. 104:345–355.
26. Brown, W. C., L. Bissey, K. S. Logan, N. C. Pedersen, J. H. Elder, and E. W.Collison. 1991. Feline immunodeficiency virus infects both CD41 andCD81 T lymphocytes. J. Virol. 65:3359–3364.
27. Brunner, D., and N. C. Pedersen. 1989. Infection of peritoneal macro-phages in vitro and in vivo with feline immunodeficiency virus. J. Virol.63:5483–5488.
28. Callanan, J. J., M. J. Hosie, and O. Jarrett. 1991. Transmission of felineimmunodeficiency virus from mother to kitten. Vet. Rec. 128:332–333.
29. Callanan, J. J., I. A. P. McCandlish, B. O’Neil, C. E. Lawrence, M. Rigby,A. M. Pacitti, and O. Jarret. 1992. Lymphosarcoma in experimentallyinduced feline immunodeficiency virus infection. Vet. Rec. 130:293–295.
30. Callanan, J. J., P. Racz, H. Thompson, and O. Jarrett. 1993. Morphologiccharacterization of the lymph node changes in feline immunodeficiencyvirus infection as an animal model of AIDS, p. 115–136. In P. Racz, N. L.Letvin, and J. C. Gluckman (ed.), Animal models of HIV and otherretroviral infections. Karger, Basel.
31. Callanan, J. J., H. Thompson, S. R. Toth, B. O’Neil, C. E. Lawrence, B.Willet, and O. Jarrett. 1992. Clinical and pathological findings in felineimmunodeficiency virus experimental infection. Vet. Immunol. Immuno-pathol. 35:3–13.
32. Chalmers, S., R. O. Schick, and J. Jeffers. 1989. Demodicosis in two catsseropositive for feline immunodeficiency virus. J. Am. Vet. Med. Assoc.194:256–257.
33. Cheng-Mayer, C., T. Shioda, and J. A. Levy. 1991. Host range, replicative,and cytopathic properties of human immunodeficiency virus type 1 aredetermined by very few amino acid changes in tat and gp120. J. Virol.65:6931–6941.
34. Cohen, J. 1993. Jitters jeopardize AIDS vaccine trials. Science 262:980–981.35. Cohen, N. D., C. N. Carter, M. A. Thomas, T. L. Lester, and A. K. Eugster.
1990. Epizootiologic association between feline immunodeficiency virusinfection and feline leukemia virus seropositivity. J. Am. Vet. Med. Assoc.197:220–225.
36. Cranage, M. P., N. Polyanskaya, B. McBride, N. Cook, L. A. E. Ashworth,M. Dennis, A. Baskerville, P. J. Greenaway, T. Corcoran, P. Kitchin, J.Rose, M. Murphey-Corb, R. C. Desrosiers, E. J. Stott, and G. H. Farrar.1993. Studies on the specificity of the vaccine effect elicited by inactivatedsimian immunodeficiency virus. AIDS Res. Hum. Retroviruses 9:13–22.
37. Cronn, R. C., K. M. Remington, B. D. Preston, and T. W. North. 1992.Inhibition of reverse transcriptase from feline immunodeficiency virus byanalogs of 29-deoxyadenosine-59-triphosphate. Biochem. Pharmacol. 44:1375–1381.
38. Cronn, R. C., J. D. Whitmer, and T. W. North. 1992. RNase H activityassociated with reverse transcriptase from feline immunodeficiency virus. J.Virol. 66:1215–1218.
39. Dalgleish, A. G. 1993. Models for HIV pathogenesis. AIDS Res. Rev.3:73–93.
40. Dandekar, S., A. M. Beebe, J. Barlough, T. Phillips, J. Elder, M. Torten,and N. Pedersen. 1992. Detection of feline immunodeficiency virus (FIV)nucleic acids in FIV-seronegative cats. J. Virol. 66:4040–4049.
40a.Davidson, M. G., J. B. Rottman, R. V. English, M. R. Lappin, and M. B.Tompkins. 1993. Feline immunodeficiency virus predisposes cats to acutegeneralized toxoplasmosis. Am. J. Pathol. 143:1486–1497.
41. Dawson, S., N. R. Smyth, M. Bennett, R. M. Gaskell, C. M. McCracken, A.Brown, and C. J. Gaskell. 1991. Effect of primary-stage feline immunode-ficiency virus infection on subsequent feline calicivirus vaccination andchallenge in cats. AIDS 5:747–750.
42. deRonde, A., J. G. Stone, P. Boers, H. Langedijk, R. Meloen, W. Hesselink,L. C. E. J. M. Keldermans, A. van Vleit, E. J. Verschoor, M. C. Horzinek,and H. F. Egberink. 1993. Antibody response in cats to the envelopeproteins of feline immunodeficiency virus: identification of an immunodom-inant neutralization domain. Virology 198:257–264.
43. Desrosiers, R. C. 1990. The simian immunodeficiency viruses. Annu. Rev.Immunol. 8:557–578.
44. Dieth, V., H. Lutz, B. Hauser, and P. Ossent. 1989. Pathologische Befundebei mit Lentiviren infizierten Katzen. Schweiz. Arch. Tierheilkd. 131:19–25.
45. Dow, S. W., M. J. Drietz, and E. A. Hoover. 1992. Feline immunodeficiencyvirus neutrotropism: evidence that astrocytes and microglia are the primarytarget cells. Vet. Immunol. Immunopathol. 35:23–35.
46. Dow, S. W., M. L. Poss, and E. A. Hoover. 1990. Feline immunodeficiencyvirus: a neurotropic lentivirus. J. Acquired Immune Defic. Syndr. 3:658–668.
47. Dua, N., G. Reubel, J. Higgins, P. F. Moore, and N. C. Pedersen. Anexperimental study of feline immunodeficiency virus infection in cats and ahistorical comparison to acute simian- and human-immunodeficiency virus
diseases. Vet. Immunol. Immunopathol., in press.48. Egberink, H., M. Borst, H. Niphuis, J. Balzarini, H. Neu, H. Schellekens, E.
De Clercq, M. Horzinek, and M. Koolen. 1990. Suppression of felineimmunodeficiency virus infection in vivo by 9-(2-phosphonomethyoxyeth-yl)adenine. Proc. Natl. Acad. Sci. USA 87:3087–3091.
49. Egberink, H. F., A. Berrocal, H. A. D. Bax, T. S. G. A. M. van den Ingh, J. H.Walter, and M. C. Horzinek. 1992. Papillomavirus associated skin lesions ina cat seropositive for feline immunodeficiency virus. Vet. Microbiol.31:117–125.
50. Egberink, H. F., J. Ederveen, R. C. Montelaro, N. C. Pedersen, M. C.Horzinek, and M. J. M. Koolen. 1990. Intracellular proteins of felineimmunodeficiency virus and their antigenic relationship with equine infec-tious anemia virus proteins. J. Gen. Virol. 71:739–743.
51. Egberink, H. F., E. J. Keldermans, M. J. M. Koolen, and M. C. Horzinek.1992. Humoral immune response to feline immunodeficiency virus in catswith experimentally induced and naturally acquired infections. Am. J. Vet.Res. 53:1133–1138.
52. Egberink, H. F., L. Keldermans, N. Schuurman, J. Stam, W. Hesselink, A.van Vliet, E. Verschoor, M. Horzinek, and A. de Ronde. 1994. Monoclonalantibodies to immunodominant and neutralizing domains of the envelopesurface protein of feline immunodeficiency virus. J. Gen. Virol. 75:889–893.
53. Egberink, H. F., H. Lutz, and M. C. Horzinek. 1991. Use of Western blotand radioimmunoprecipitation for diagnosis of feline leukemia and felineimmunodeficiency virus infections. J. Am. Vet. Med. Assoc. 199:1339–1342.
54. Elder, J. H., D. A. Lerner, C. S. Hasselkus-Light, D. J. Fontenot, E. Hunter,P. A. Luciw, R. C. Montelaro, and T. M. Phillips. 1992. Distinct subsets ofretroviruses encode dUTPase. J. Virol. 66:1791–1794.
55. Elder, J. H., M. Schnolzer, C. S. Hasselkus-Light, M. Henson, D. A. Lerner,T. R. Phillips, P. C. Wagaman, and S. B. H. Kent. 1993. Identification ofproteolytic processing sites within the Gag and Pol polyproteins of felineimmunodeficiency virus. J. Virol. 67:1869–1876.
56. English, R. V., C. M. Johnson, D. H. Gebhard, and M. B. Tompkins. 1993.In vivo lymphocyte tropism of feline immunodeficiency virus. J. Virol.67:5175–5186.
57. Fauci, A. S. 1993. Multifactorial nature of human immunodeficiency virusdisease: implications for therapy. Science 262:1011–1018.
58. Feder, B. M., and A. Hurvitz. 1990. Feline immunodeficiency virus infectionin 100 cats and association with lymphoma. J. Vet. Intern. Med. 4:110.
59. Fevereiro, M., C. Roneker, and F. De Noronha. 1991. Antibody response toreverse transcriptase in cats infected with feline immunodeficiency virus.Viral. Immunol. 4:225–235.
60. Fevereiro, M., C. Roneker, and F. De Noronha. 1993. Enhanced neutral-ization of feline immunodeficiency virus by complement virus lysis. Vet.Immunol. Immunopathol. 36:191–206.
61. Fevereiro, M., C. Roneker, A. Laufs, L. Tavares, and F. de Noronha. 1991.Characterization of two monoclonal antibodies against feline immunodefi-ciency virus gag gene products and their application in an assay to evaluateneutralizing antibody activity. J. Gen. Virol. 72:617–622.
62. Fleming, E. J., D. L. McCaw, J. A. Smith, G. M. Buening, and C. Johnson.1991. Clinical, hematologic, and survival data from cats infected with felineimmunodeficiency virus: 42 cases (1983–1988). J. Am. Vet. Med. Assoc.199:913–916.
63. Flynn, J. N., C. A. Cannon, C. E. Lawrence, and O. Jarrett. 1994. PolyclonalB-cell activation in cats infected with feline immunodeficiency virus.Immunology 81:626–630.
64. Fontenot, J. D., E. A. Hoover, J. H. Elder, and R. C. Montelaro. 1992.Evaluation of feline immunodeficiency virus and feline leukemia virustransmembrane peptides for serological diagnosis. J. Clin. Microbiol.30:1885–1890.
65. Fujimoto, S., M. Masuda, E. Ono, H. Kida, and Y. Shimizu. 1993.Establishment of a feline T-lymphoblastoid cell line susceptible to felineimmunodeficiency viruses. Microbiol. Immunol. 37:415–418.
66. Furuya, T., A. Hasegawa, T. Miyazawa, K. Miki, and T. Mikami. 1992.Detection of anti-gag antibodies of feline immunodeficiency virus in cat seraby enzyme-linked immunosorbent assay. Arch. Virol. 124:355–361.
67. Gardner, M. B. 1991. Simian and feline immunodeficiency viruses: animallentivirus models for evaluation of AIDS vaccines and antiviral agents.Antiviral Res. 15:267–286.
68. George, J. W., N. C. Pedersen, and J. Higgins. 1993. The effect of age on thecourse of experimental feline immunodeficiency virus infection in cats.AIDS Res. Hum. Retroviruses 9:897–905.
69. Greene, W. K., J. Meers, B. Chadwick, P. R. Carnegie, and W. F. Robinson.1993. Nucleotide sequence of Australian isolates of the feline immunode-ficiency virus: comparison with other feline lentiviruses. Arch. Virol.132:369–379.
70. Greene, W. K., J. Meers, G. del Fierro, P. R. Carnegie, and W. F. Robinson.1993. Extensive sequence variation of feline immunodeficiency virus envgenes in isolates from naturally infected cats. Arch. Virol. 133:51–62.
71. Grindem, C. B., W. T. Corbett, B. E. Ammerman, and M. T. Tomkins. 1989.Seroepidemiologic survey of feline immunodeficiency virus infection in catsof Wake County, North Carolina. J. Am. Vet. Med. Assoc. 194:226–228.
72. Gruffydd-Jones, T. J., C. D. Hopper, D. A. Harbour, and H. Lutz. 1988.
Serological evidence of feline immunodeficiency virus infection in UK catsfrom 1975–76. Vet. Rec. 123:569–570.
73. Hafkemeyer, P., K. Neftel, R. Hobi, A. Pfaltz, H. Lutz, K. Luthi, F. Focher,S. Spadari, and U. Hubscher. 1991. HP 0.35, a cephalosporin degradationproduct is a specific inhibitor of lentiviral RNAses H. Nucleic Acids Res.9:4059–4065.
74. Hanlon, M. A., J. M. Marr, K. A. Hayes, L. E. Mathes, P. C. Stromberg, S.Ringler, S. Krakowka, and L. J. Lafrado. 1993. Loss of neutrophil andnatural killer cell function following feline immunodeficiency virus infec-tion. Viral Immunol. 6:119–124.
75. Hara, Y., T. Ishida, H. Ejima, M. Tagawa, S. Motoyoshi, I. Tomoda, M.Shimizu, and K. Shichinohe. 1990. Decrease in mitogen-induced lympho-cyte proliferative responses in cats infected with feline immunodeficiencyvirus. Jpn. J. Vet. Sci. 52:573–579.
76. Harbour, D., S. Bishop, C. Whiting, T. Gruffidd-Jones, and C. Stokes. 1993.Mucosal infection and immunisation studies with FIV in the cat, p. 50. InAbstracts of the International Symposium on Feline Retrovirus Research,1993. North Carolina State University, Raleigh.
77. Harbour, D. A. 1992. Feline immunodeficiency virus infection as a modelfor HIV infection in man. Rev. Med. Virol. 2:43–49.
78. Harbour, D. A., P. D. Williams, T. J. Gruffydd-Jones, J. Burbridge, andG. R. Pearson. 1988. Isolation of a T-lymphotropic lentivirus from apersistently leucopenic domestic cat. Vet. Rec. 122:84–86.
79. Hartmann, K., A. Donath, B. Beer, H. F. Egberink, M. C. Horzinek, H.Lutz, G. Hoffmann-Fezer, I. Thum, and S. Thefeld. 1992. Use of twovirustatica (AZT, PMEA) in the treatment of FIV and of FeLV seroposi-tive cats with clinical symptoms. Vet. Immunol. Immunopathol. 35:167–175.
80. Haynes, B. F. 1993. Scientific and social issues of human immunodeficiencyvirus vaccine development. Science 260:1279–1286.
81. Heidel, J. R., J. P. Dubey, L. L. Blythe, L. L. Walker, J. R. Duimstra, andJ. S. Jordan. 1990. Myelitis in a cat infected with Toxoplasma gondii andfeline immunodeficiency virus. J. Am. Vet. Med. Assoc. 196:316–318.
82. Hoffmann-Fezer, G., J. Thum, C. Ackley, M. Herbold, J. Mysliwietz, S.Thefeld, K. Hartmann, and W. Kraft. 1992. Decline in CD41 cell numbersin cats with naturally acquired feline immunodeficiency virus infection. J.Virol. 66:1484–1488.
83. Hohdatsu, T., R. Pu, B. A. Torres, S. Trujillo, M. B. Gardner, and J. K.Yamamoto. 1993. Passive antibody protection of cats against feline immu-nodeficiency virus infection. J. Virol. 67:2344–2348.
84. Hohdatsu, T., M. Yamada, M. Okada, M. Fukasawa, K. Watanabe, T.Ogasawara, M. Takagi, C. Aizawa, M. Hayami, and H. Koyama. 1992.Detection of feline immunodeficiency virus proviral DNA in peripheralblood lymphocytes by the polymerase chain reaction. Vet. Microbiol.30:113–123.
85. Hopper, C. D., A. H. Sparkes, T. J. Gruffydd-Jones, S. M. Crispin, P. Muir,D. A. Harbour, and C. R. Stokes. 1989. Clinical and laboratory findings incats infected with feline immunodeficiency virus. Vet. Rec. 125:341–346.
86. Hosie, M. J. 1994. The development of a vaccine against feline immuno-deficiency virus. Br. Vet. J. 150:25–39.
87. Hosie, M. J., and O. Jarrett. 1990. Serological responses of cats to felineimmunodeficiency virus. AIDS 4:215–220.
88. Hosie, M. J., R. Osborne, G. Reid, J. C. Neil, and O. Jarrett. 1992.Enhancement after feline immunodeficiency virus vaccination. Vet. Immu-nol. Immunopathol. 35:191–197.
89. Hosie, M. J., R. Osborne, B. J. Willett, F. J. Thompson, J. C. Neil, and O.Jarrett. 1993. Protection from FIV infection using inactivated virus vac-cines, p. 50. In Abstracts of the International Symposium on FelineRetrovirus Research, 1993. North Carolina State University, Raleigh.
90. Hosie, M. J., C. Robertson, and O. Jarrett. 1989. Prevalence of felineleukaemia virus and antibodies to feline immunodeficiency virus in cats inthe United Kingdom. Vet. Rec. 128:293–297.
91. Hosie, M. J., B. J. Willett, T. H. Dunsford, O. Jarrett, and J. C. Neil. 1993.A monoclonal antibody which blocks infection with feline immunodefi-ciency virus identifies a possible non-CD4 receptor. J. Virol. 67:1667–1671.
92. Hu, S.-L., K. Abrams, G. N. Barber, P. Moran, J. M. Zarling, A. J. Langlois,L. Kuller, W. R. Morton, and R. E. Benveniste. 1992. Protection ofmacaques against SIV infection by subunit vaccines of SIV envelopeglycoprotein gp160. Science 255:456–459.
93. Hurtrel, M., J.-P. Ganiere, J.-F. Guelfi, L. Chakrabarti, M.-A. Maire, F.Gray, L. Montagnier, and B. Hurtrel. 1992. Comparison of early and latefeline immunodeficiency virus encephalopathies. AIDS 6:399–406.
94. Hutson, C. A., B. A. Rideout, and N. C. Pedersen. 1991. Neoplasiaassociated with feline immunodeficiency virus infection in cats of SouthernCalifornia. J. Am. Vet. Med. Assoc. 199:1357–1362.
95. Ishida, T., A. Taniguchi, S. Matsumura, T. Washizu, and I. Tomoda. 1992.Long-term clinical observations on feline immunodeficiency virus infectedasymptomatic carriers. Vet. Immunol. Immunopathol. 35:15–22.
96. Ishida, T., and I. Tomoda. 1990. Clinical staging of feline immunodeficiencyvirus infection. Jpn. J. Vet. Sci. 52:645–648.
97. Ishida, T., T. Washizu, K. Toriyabe, and S. Motoyoshi. 1988. Detection offeline T-lymphotropic lentivirus (FTLV) infection in Japanese domesticcats. Jpn. J. Vet. Sci. 50:39–44.
98. Ishida, T., T. Washizu, K. Toriyabe, S. Motoyoshi, I. Tomoda, and N. C.Pedersen. 1989. Feline immunodeficiency virus infection in cats of Japan. J.Am. Vet. Med. Assoc. 194:221–225.
99. Johnston, M. I., and D. F. Hoth. 1993. Present status and future prospectsfor HIV therapies. Science 260:1286–1293.
100. Kawaguchi, Y., K. Maeda, Y. Tohya, T. Furuya, T. Miyazawa, T. Horimoto,J. Norinime, C. Kai, and T. Mikami. 1992. Replicative difference inearly-passage feline brain cells among feline immunodeficiency virus iso-lates. Arch. Virol. 125:347–354.
101. Kawaguchi, Y., T. Miyazawa, T. Horimoto, S. I. Itagaki, M. Fukasawa, E.Takahashi, and T. Mikami. 1991. Activation of feline immunodeficiencyvirus long terminal repeat by feline herpesvirus type 1. Virology 184:449–454.
102. Kawaguchi, Y., J. Norimine, T. Miyazawa, C. Kai, and T. Mikami. 1992.Sequences within the feline immunodeficiency virus long terminal repeatthat regulate gene expression and respond to activation by feline herpesvi-rus type 1. Virology 190:465–468.
103. Kindt, T. J., V. M. Hirsch, P. R. Johnson, and S. Sawasdikosol. 1992.Animal models for acquired immunodeficiency syndrome. Adv. Immunol.52:425–474.
104. Kiyomasu, T., T. Miyazawa, T. Furuya, R. Shibata, H. Sakai, J.-I. Sakuragi,M. Fukasawa, N. Maki, A. Hasegawa, T. Mikami, and A. Adachi. 1991.Identification of feline immunodeficiency virus rev gene activity. J. Virol.65:4539–4542.
105. Knowles, J. O., R. M. Gaskell, C. J. Gaskell, C. E. Harvey, and H. Lutz.1989. Prevalence of feline calicivirus, feline leukaemia virus and antibodiesto FIV in cats with chronic stomatitis. Vet. Rec. 124:336–338.
106. Lackner, A. A., S. Dandekar, and M. B. Gardner. 1991. Neurobiology ofsimian and feline immunodeficiency virus infections. Brain Pathol. 1:201–212.
107. Lafrado, L. J., M. Podell, S. Krakowka, K. A. Hayes, M. A. Hanlon, andL. E. Mathes. 1993. FIV: a model for retrovirus-induced pathogenesis.AIDS Res. Rev. 3:115–150.
108. Lappin, M. R., P. W. Gasper, B. J. Rose, and C. C. Powell. 1992. Effect ofprimary phase feline immunodeficiency virus infection on cats with chronictoxoplasmosis. Vet. Immunol. Immunopathol. 35:121–132.
109. Lawrence, C. E., J. J. Callanan, and O. Jarrett. 1992. Decreased mitogenresponsiveness and elevated tumor necrosis factor production in catsshortly after feline immunodeficiency virus infection. Vet. Immunol. Im-munopathol. 35:51–59.
110. Lehmann, R., M. Franchini, A. Aubert, C. Wolfensberger, J. Cronier, andH. Lutz. 1992. Vaccination of cats experimentally infected with felineimmunodeficiency virus, using a recombinant feline leukemia virus vaccine.J. Am. Vet. Med. Assoc. 199:1446–1452.
111. Lehmann, R., H. Joller, B. L. Haagmans, and H. Lutz. 1992. Tumornecrosis factor a levels in cats experimentally infected with feline immu-nodeficiency virus: effects of immunization and feline leukemia virusinfection. Vet. Immunol. Immunopathol. 35:61–69.
112. Lehmann, R., B. von Beust, E. Niederer, M. A. Condrau, W. Fierz, A.Aubert, C. D. Ackley, M. D. Cooper, M. B. Tompkins, and H. Lutz. 1992.Immunization-induced decrease of the CD41:CD81 ratio in cats experi-mentally infected with feline immunodeficiency virus. Vet. Immunol.Immunopathol. 35:199–214.
113. Levy, J. A. 1993. Pathogenesis of human immunodeficiency virus infection.Microbiol. Rev. 57:183–289.
115. Lin, D.-S., and D. D. Bowman. 1992. Macrophage functions in catsexperimentally infected with feline immunodeficiency virus and Toxoplasmagondii. Vet. Immunol. Immunopathol. 33:69–78.
116. Lin, D.-S., D. D. Bowman, and R. H. Jacobson. 1992. Immunologicalchanges in cats with concurrent Toxoplasma gondii and feline immunode-ficiency virus infections. J. Clin. Microbiol. 30:17–24.
117. Lin, D.-S., D. D. Bowman, R. H. Jacobson, M. C. Barr, M. Fevereiro, J. R.Williams, F. M. O. Noronha, F. W. Scott, and R. J. Avery. 1990. Suppressionof lymphocyte blastogenesis to mitogens in cats experimentally infectedwith feline immunodeficiency virus. Vet. Immunol. Immunopathol. 26:183–189.
118. Linenberger, M., A. Beebe, S. Dandekar, N. Pedersen, and J. Abkowitz.1993. Alterations in hematopoiesis during primary infection with FIV, p. 8.In Abstracts of the International Symposium on Feline Retrovirus Re-search, 1993. North Carolina State University, Raleigh.
119. Lombardi, S., M. Bendinelli, and T. Garzelli. 1993. Detection of B epitopeson the p24 gag protein of feline immunodeficiency virus by monoclonalantibodies. AIDS Res. Hum. Retroviruses 9:141–146.
120. Lombardi, S., C. Garzelli, C. La Rosa, L. Zaccaro, S. Specter, G. Malvaldi,F. Tozzini, F. Esposito, and M. Bendinelli. 1993. Identification of a linearneutralization site within the third variable region of feline immunodefi-ciency virus envelope. J. Virol. 67:4742–4749.
121. Lombardi, S., C. Garzelli, S. Specter, G. Malvaldi, F. Tozzini, F. Esposito,and M. Bendinelli. 1993. Identification of a linear neutralization epitope inthe third variable region of FIV envelope, p. 30. In Abstracts of the
International Symposium on Feline Retrovirus Research, 1993. NorthCarolina State University, Raleigh.
122. Lombardi, S., A. Poli, C. Massi, F. Abramo, L. Zaccaro, A. Bazzichi, G.Malvaldi, M. Bendinelli, and C. Garzelli. 1994. Detection of feline immu-nodeficiency virus p24 antigen and p24-specific antibodies by monoclonalantibody-based assays. J. Virol. Methods 46:287–301.
123. Lutz, H. 1990. Feline retroviruses: a brief review. Vet. Microbiol. 23:131–146.
124. Lutz, H., E. Isenbugel, R. Lehmann, R. H. Sabapara, and C. Wolfensberger.1992. Retrovirus infections in non-domestic felids: serological studies andattempts to isolate a lentivirus. Vet. Immunol. Immunopathol. 35:215–224.
125. Lutz, H., R. Lehmann, K. Bauer-Pham, F. Tozzini, M. Bendinelli, G.Reubel, A. Aubert, and E. Young. 1993. Vaccination of cats with recombi-nant FIV env-gene products, p. 48. In Abstracts of the InternationalSymposium on Feline Retrovirus Research, 1993. North Carolina StateUniversity, Raleigh.
126. Lutz, H., R. Lehmann, G. Winkler, B. Kottwitz, A. Dittmer, C. Wolfens-berger, and P. Arnold. 1990. Das feline Immunschwachevirus in derSchweiz: Klinik und Epidemiologie im Vergleich mit dem Leukamie-unddem Coronavirus. Schweiz. Arch. Tierheilkd. 132:217–225.
127. Magnani, M., L. Rossi, A. Fraternale, L. Silvotti, F. Quintavalla, G.Piedimonte, D. Matteucci, F. Baldinotti, and M. Bendinelli. FIV infectionof macrophages: in vitro and in vivo inhibition by dideoxycytidine 59-triphosphate. Vet. Immunol. Immunopathol., in press.
128. Magnani, M., L. Rossi, A. Fraternale, L. Silvotti, F. Quintavalla, G.Piedimonte, D. Matteucci, F. Baldinotti, and M. Bendinelli. 1994. Felineimmunodeficiency virus infection of macrophages: in vitro and in vivoinhibition by ddCTP-loaded erythrocytes. AIDS Res. Hum. Retroviruses9:1179–1186.
129. Maki, N., T. Miyazawa, M. Fukasawa, A. Hasegawa, M. Hayami, K. Miki,and T. Mikami. 1992. Molecular characterization and heterogeneity offeline immunodeficiency virus isolates. Arch. Virol. 123:29–45.
130. Mancianti, F., C. Giannelli, M. Bendinelli, and A. Poli. 1992. Mycologicalfindings in feline immunodeficiency virus-infected cats. J. Med. Vet. Mycol.30:257–259.
131. Mandell, C. P., E. E. Sparger, N. C. Pedersen, and N. C. Jain. 1992.Long-term hematological changes in cats experimentally infected withfeline immunodeficiency virus (FIV). Comp. Hematol. Int. 2:8–17.
132. Marsilio, F., M. Tempesta, M. G. Pennisi, and D. Buonavoglia. 1990.Experimental infection of cats with feline immunodeficiency virus: furthertrials. Atti Soc. Ital. Sci. Vet. 44:813–817.
133. Matsumura, S., T. Ishida, T. Washizu, I. Tomoda, S. Nagata, J. Chiba, andT. Kurata. 1993. Pathologic features of acquired immunodeficiency-likesyndrome in cats experimentally infected with feline immunodeficiencyvirus. J. Vet. Med. Sci. 55:387–394.
134. Matsuo, K., Y. Nishino, T. Kimura, R. Yamaguchi, A. Yamazaki, T.Mikami, and K. Ikuta. 1992. Highly conserved epitope domain in majorcore protein p24 is structurally similar among human, simian and felineimmunodeficiency viruses. J. Gen. Virol. 73:2445–2450.
135. Matteucci, D., F. Baldinotti, P. Mazzetti, M. Pistello, P. Bandecchi, R.Ghilarducci, A. Poli, F. Tozzini, and M. Bendinelli. 1993. Detection offeline immunodeficiency virus in saliva and plasma by cultivation andpolymerase chain reaction. J. Clin. Microbiol. 31:494–501.
136. Matteucci, D., P. Mazzetti, F. Baldinotti, L. Zaccaro, and M. Bendinelli.The feline lymphoid cell line MBM and its use for feline immunodeficiencyvirus isolation and quantitation. Vet. Immunol. Immunopathol., in press.
137. Meers, J., G. M. del Fierro, R. B. Cope, H. S. Park, W. K. Greene, and W. F.Robinson. 1993. Feline immunodeficiency virus infection: plasma, but notperipheral blood mononuclear cell virus titer is influenced by zidovudineand cyclosporine. Arch. Virol. 132:67–81.
138. Meers, J., W. F. Robinson, G. M. del Fierro, M. A. Scoones, and M. A.Lawson. 1992. Feline immunodeficiency virus: quantification in peripheralblood mononuclear cells and isolation from plasma of infected cats. Arch.Virol. 127:233–243.
139. Mermer, B., P. Hillman, R. Harris, T. Krogman, Q. Tonelli, W. Palin, andP. Andersen. 1992. A recombinant-based feline immunodeficiency virusantibody enzyme-linked immunosorbent assay. Vet. Immunol. Immuno-pathol. 35:133–141.
140. Miyazawa, T., M. Fukasawa, A. Hasegawa, N. Maki, K. Ikuta, E. Taka-hashi, M. Hayami, and T. Mikami. 1991. Molecular cloning of a novelisolate of feline immunodeficiency virus biologically and genetically differ-ent from the original U.S. isolate. J. Virol. 65:1572–1577.
141. Miyazawa, T., T. Furuya, S.-I. Itagaki, Y. Tohya, K. Nakano, E. Takahashi,and T. Mikami. 1989. Preliminary comparisons of the biological propertiesof two strains of feline immunodeficiency virus (FIV) isolated in Japan withFIV Petaluma strain isolated in the United States. Arch. Virol. 108:59–68.
142. Miyazawa, T., Y. Kawaguchi, M. Kohmoto, J.-I. Sakuragi, A. Adachi, M.Fukasawa, and T. Mikami. 1992. Production of feline immunodeficiencyvirus in feline and non-feline non-lymphoid cell lines by transfection of aninfectious molecular clone. J. Gen. Virol. 73:1543–1546.
143. Miyazawa, T., M. Kohmoto, Y. Kawaguchi, K. Tomonaga, T. Toyosaki, K.Ikuta, A. Adachi, and T. Mikami. 1993. The AP-1 binding site in the feline
immunodeficiency virus long terminal repeat is not required for virusreplication in feline T lymphocytes. J. Gen. Virol. 74:1573–1580.
144. Mochizuki, M., M. Akuzawa, and H. Nagatomo. 1990. Serological survey ofthe iriomote cat (Felis iriomotensis) in Japan. J. Wildl. Dis. 26:236–245.
145. Moench, T. R., K. J. Whaley, T. D. Mandrell, B. D. Bishop, C. J. Witt, andR. A. Cone. 1993. The cat/feline immunodeficiency virus model for trans-mucosal transmission of AIDS: nonoxynol-9 contraceptive jelly blockstransmission by an infected cell inoculum. AIDS 7:797–802.
146. Montefiori, D. C. 1993. In vitro correlates of HIV and SIV humoralimmunity and infection enhancement. AIDS Res. Rev. 3:161–183.
148. Moraillon, A., F. Barre-Sinoussi, A. Parodi, R. Moraillon, and C. Dauguet.1992. In vitro properties and experimental pathogenic effect of three strainsof feline immunodeficiency virus isolated from cats with terminal disease.Vet. Microbiol. 31:41–54.
149. Morikawa, S., and D. H. L. Bishop. 1992. Identification and analysis of thegag-pol ribosomal frameshift site of feline immunodeficiency virus. Virology186:389–397.
150. Morikawa, S., T. F. Booth, and D. H. L. Bishop. 1991. Analyses of therequirements for the synthesis of virus-like particles by feline immunode-ficiency virus gag using baculovirus vectors. Virology 183:288–297.
151. Morikawa, S., H. Lutz, A. Aubert, and D. H. L. Bishop. 1991. Identificationof conserved and variable regions in the envelope glycoprotein sequences oftwo feline immunodeficiency viruses isolated in Zurich, Switzerland. VirusRes. 21:53–63.
152. Mtambo, M. M. A., A. S. Nash, D. A. Blewett, H. V. Smith, and S. Wright.1991. Cryptosporidium infection in cats: prevalence of infection in domesticand feral cats in the Glasgow area. Vet. Rec. 129:502–504.
153. Nara, P. L., R. R. Garrity, and J. Goudsmit. 1991. Neutralization of HIV-1:a paradox of humoral proportions. FASEB J. 5:2437–2455.
154. Nishino, T., K. Ohki, T. Kimura, S. Morikawa, T. Mikami, and K. Ikuta.1992. Major core proteins, p24s, of human, simian, and feline immunode-ficiency viruses are poorly expressed on the surface of the virus-infectedcells. Vaccine 10:677–683.
155. Norimine, J., T. Miyazawa, Y. Kawaguchi, K. Tomonaga, Y.-S. Shin, T.Toyosaki, M. Kohmoto, M. Niikura, Y. Tohya, and T. Mikami. 1993. FelineCD4 molecules expressed on feline non-lymphoid cell lines are not enoughfor productive infection of highly lymphotropic feline immunodeficiencyvirus isolates. Arch. Virol. 130:171–178.
156. North, T. W., R. C. Cronn, K. M. Remington, and R. T. Tandberg. 1990.Direct comparisons of inhibitor sensitivities of reverse transcriptases fromfeline and human immunodeficiency viruses. Antimicrob. Agents Che-mother. 34:1505–1507.
157. North, T. W., R. C. Cronn, K. M. Remington, R. T. Tandberg, and R. C.Judd. 1990. Characterization of reverse transcriptase from feline immuno-deficiency virus. J. Biol. Chem. 265:5121–5128.
158. North, T. W., G. L. T. North, and N. C. Pedersen. 1989. Feline immuno-deficiency virus, a model for reverse transcriptase-targeted chemotherapyfor acquired immune deficiency syndrome. Antimicrob. Agents Chemother.33:915–919.
159. Novotney, C., R. V. English, J. Housman, M. G. Davidson, M. P. Nasisse,C.-R. Jeng, W. C. Davis, and M. B. Tompkins. 1990. Lymphocyte popula-tion changes in cats naturally infected with feline immunodeficiency virus.AIDS 4:1213–1218.
160. O’Connor, T. P., Jr., S. Tanguay, R. Steinman, R. Smith, M. C. Barr, J. K.Yamamoto, N. C. Pedersen, P. R. Andersen, and Q. J. Tonelli. 1989.Development and evaluation of immunoassay for detection of antibodies tothe feline T-lymphotropic lentivirus (feline immunodeficiency virus). J.Clin. Microbiol. 27:474–479.
161. O’Connor, T. P., Q. J. Tonelli, and J. M. Scarlett. 1991. Report of thenational FeLV/FIV awareness project. J. Am. Vet. Med. Assoc. 199:1348–1353.
162. Ohno, K., T. Nakano, Y. Matsumoto, T. Watari, R. Goitsuka, H. Nakayama,H. Tsujimoto, and A. Hasegawa. 1993. Apoptosis induced by tumor necrosisfactor in cells chronically infected with feline immunodeficiency virus. J.Virol. 67:2429–2433.
163. Ohno, K., T. Watari, R. Goitsuka, H. Tsujimoto, and A. Hasegawa. 1992.Altered surface antigen expression on peripheral blood mononuclear cellsin cats infected with feline immunodeficiency virus. J. Vet. Med. Sci.54:517–522.
164. Olmsted, R. A., A. K. Barnes, J. K. Yamamoto, V. M. Hirsch, R. H. Purcell,and P. R. Johnson. 1989. Molecular cloning of feline immunodeficiencyvirus. 86:2248–2252.
165. Olmsted, R. A., V. M. Hirsch, R. H. Purcell, and P. R. Johnson. 1989.Nucleotide sequence analysis of feline immunodeficiency virus: genomeorganization and relationship to other lentiviruses. Proc. Natl. Acad. Sci.USA 86:8088–8092.
166. Olmsted, R. A., R. Langley, M. E. Roelke, R. G. Goeken, D. Adger-Johnson,J. P. Goff, J. P. Albert, C. Packer, M. K. Laurenson, T. M. Caro, L.Scheepers, D. E. Wildt, M. Bush, J. S. Martenson, and S. J. O’Brien. 1992.Wordwide prevalence of lentivirus infection in wild feline species: epide-
miologic and phylogenetic aspects. J. Virol. 66:6008–6018.167. O’Neil, S. A., M. R. Lappin, J. S. Reif, A. Marks, and C. E. Greene. 1991.
Clinical and epidemiological aspects of feline immunodeficiency virus andToxoplasma gondii coinfections in cats. J. Am. Anim. Hosp. Assoc. 27:211–220.
168. Orentas, R. J., and J. E. K. Hildreth. 1993. Association of host cell surfaceadhesion receptors and other membrane proteins with HIV and SIV. AIDSRes. Hum. Retroviruses 9:1157–1165.
169. Pancino, G., C. Chappey, W. Saurin, and P. Sonigo. 1993. B epitopes andselection pressures in feline immunodeficiency virus envelope glycopro-teins. J. Virol. 67:664–672.
170. Pancino, G., I. Fossati, C. Chappey, S. Castelot, B. Hurtrel, A. Moraillon,D. Klatzmann, and P. Sonigo. 1993. Structure and variations of felineimmunodeficiency virus envelope glycoproteins. Virology 192:659–662.
171. Pearson, L. D., M. L. Poss, and J. C. Demartini. 1989. Animal lentivirusvaccines: problems and prospects. Vet. Immunol. Immunopathol. 20:183–212.
172. Pedersen, N. C. 1993. Feline immunodeficiency virus infection. In J. A. Levy(ed.), The retroviruses, vol. 2. Plenum Press, New York.
173. Pedersen, N. C., and J. E. Barlough. 1991. Clinical overview of felineimmunodeficiency virus. J. Am. Vet. Med. Assoc. 199:1298–1305.
174. Pedersen, N. C., E. W. Ho, M. L. Brown, and J. K. Yamamoto. 1987.Isolation of a T-lymphotropic virus from domestic cats with an immunod-eficiency-like syndrome. Science 235:790–793.
175. Pedersen, N. C., M. Torten, B. Rideout, E. Sparger, T. Tonachini, P. A.Luciw, C. Ackley, N. Levy, and J. K. Yamamoto. 1990. Feline leukemia virusinfection as a potentiating cofactor for the primary and secondary stages ofexperimentally induced feline immunodeficiency virus infection. J. Virol.64:598–606.
176. Pedersen, N. C., J. K. Yamamoto, T. Ishida, and H. Hansen. 1989. Felineimmunodeficiency virus infection. Vet. Immunol. Immunopathol. 21:111–129.
177. Phillips, T. R., C. Lamont, D. A. M. Konings, B. L. Shacklett, C. A.Hamson, P. A. Luciw, and J. H. Elder. 1992. Identification of the revtransactivation and rev-responsive elements of feline immunodeficiencyvirus. J. Virol. 66:5464–5471.
178. Phillips, T. R., R. L. Talbott, C. Lamont, S. Muir, K. Lovelace, and J. H.Elder. 1990. Comparison of two host cell range variants of feline immuno-deficiency virus. J. Virol. 64:4605–4613.
179. Philpott, M. S., J. P. Ebner, and E. A. Hoover. 1992. Evaluation of9-(2-phosphonylmethoxyethyl)adenine therapy for feline immunodeficiencyvirus using a quantitative polymerase chain reaction. Vet. Immunol.Immunopathol. 35:155–166.
180. Pistello, M., S. Menzo, M. Giorgi, L. Da Prato, G. Cammarota, M.Clementi, and M. Bendinelli. 1994. Competitive polymerase chain reactionfor quantitating feline immunodeficiency virus load in infected cat tissues.Mol. Cell. Probes 8:229–234.
181. Podell, M., M. Oglesbee, L. Mathes, S. Krakowka, R. Olmstead, and L.Lafrado. 1993. AIDS-associated encephalopathy with experimental felineimmunodeficiency virus infection. J. Acquired Immune Defic. Syndr.6:678–771.
182. Poli, A., F. Abramo, F. Baldinotti, M. Pistello, L. Da Prato, and M.Bendinelli. 1994. Malignant lymphoma associated with experimentallyinduced feline immunodeficiency virus infection. J. Comp. Pathol. 110:319–328.
183. Poli, A., F. Abramo, P. Cavicchio, P. Bandecchi, E. Ghelardi, and M.Pistello. Retroviral infection in a lion: a clinical, pathological and virolog-ical study. J. Wildl. Dis., in press.
184. Poli, A., F. Abramo, D. Matteucci, F. Baldinotti, M. Pistello, S. Lombardi,P. Barsotti, and M. Bendinelli. Renal involvement in feline immunodefi-ciency virus: p24 antigen detection, virus isolation and PCR analysis. Vet.Immunol. Immunopathol., in press.
185. Poli, A., F. Abramo, E. Taccini, G. Guidi, P. Barsotti, M. Bendinelli, and G.Malvaldi. 1993. Renal involvement in feline immunodeficiency virus infec-tion: a clinicopathological study. Nephron 64:282–288.
186. Poli, A., C. Giannelli, M. Pistello, L. Zaccaro, M. Pieracci, M. Bendinelli,and G. Malvaldi. 1992. Detection of salivary antibodies in cats infected withfeline immunodeficiency virus. J. Clin. Microbiol. 30:2038–2041.
187. Poss, M. L., S. W. Dow, and E. A. Hoover. 1992. Cell-specific envelopeglycosylation distinguishes FIV glycoproteins produced in cytopathicallyand noncytopathically infected cells. Virology 188:25–32.
188. Pu, R., and J. Yamamoto. 1993. Humoral immunity in FIV vaccineprotection, p. 48. In Abstracts of the International Symposium on FelineRetrovirus Research, 1993. North Carolina State University, Raleigh.
189. Reid, G., M. A. Rigby, M. McDonald, M. J. Hosie, J. C. Neil, and O. Jarrett.1991. Immunodiagnosis of feline immunodeficiency virus infection usingrecombinant viral p17 and p24. AIDS 5:1477–1483.
190. Remington, K. M., B. Chesebro, K. Wehrly, N. C. Pedersen, and T. W.North. 1991. Mutants of feline immunodeficiency virus resistant to 39-azido-39-deoxythymidine. J. Virol. 65:308–312.
191. Remington, K. M., Y.-Q. Zhu, T. R. Phillips, and T. W. North. 1994. Rapidphenotypic reversion of zidovudine-resistant feline immunodeficiency virus
without loss of drug-resistant reverse transcriptase. J. Virol. 68:632–637.192. Reubel, G., G. Dean, J. Barlough, and N. Pedersen. 1993. Immunizations
and incidental infections do not accellerate immunologic abnormalitiesoccurring in experimentally FIV infected cats, p. 16. In Abstracts of theInternational Symposium on Feline Retrovirus Research, 1993. NorthCarolina State University, Raleigh.
193. Reubel, G. H., J. W. George, J. E. Barlough, J. Higgins, C. K. Grant, andN. C. Pedersen. 1992. Interaction of acute feline herpesvirus-1 and chronicfeline immunodeficiency virus infections in experimentally infected specificpathogen free cats. Vet. Immunol. Immunopathol. 35:95–119.
194. Rideout, B. A., L. J. Lowenstine, C. A. Hutson, P. F. Moore, and N. C.Pedersen. 1992. Characterization of morphologic changes and lymphocytesubset distribution in lymph nodes from cats with naturally acquired felineimmunodeficiency virus infection. Vet. Pathol. 29:391–399.
195. Rideout, B. A., P. F. Moore, and N. C. Pedersen. 1992. Persistent upregu-lation of MHC class II antigen expression on T-lymphocytes from catsexperimentally infected with feline immunodeficiency virus. Vet. Immunol.Immunopathol. 35:71–81.
196. Rideout, D., R. Schinazi, C. D. Pauza, K. Lovelace, L.-C. Chiang, T.Calogeropoulou, M. McCarthy, and J. H. Elder. 1993. Derivatives of4-amino-3,6-disulfonato-1, 8-naphthalimide inhibit reverse transcriptaseand suppress human and feline immunodeficiency virus expression incultured cells. J. Cell. Biochem. 51:446–457.
197. Rigby, M. A., E. C. Holmes, M. Pistello, A. Mackay, A. J. Leigh Brown, andJ. C. Neil. 1993. Evolution of structural proteins of feline immunodeficiencyvirus: molecular epidemiology and evidence of selection for change. J. Gen.Virol. 74:425–436.
198. Rimstad, E., and K. Ueland. 1992. Detection of feline immunodeficiencyvirus by a nested polymerase chain reaction. J. Virol. Methods 36:239–248.
199. Salk, J., P. A. Bretscher, P. L. Salk, M. Clerici, and G. M. Shearer. 1993. Astrategy for prophylactic vaccination against HIV. Science 260:1270–1272.
200. Saxinger, W. C., P. H. Levine, A. G. Dean, G. de The, G. Lange-Wantzin, J.Moghissi, F. Laurent, M. Hoh, M. G. Sarngadharan, and R. C. Gallo. 1985.Evidence of exposure to HTLV-III in Uganda before 1973. Science227:1036–1038.
201. Shelton, G. H., J. L. Abkowitz, M. L. Linenberger, R. G. Russell, and C. K.Grant. 1989. Chronic leukopenia associated with feline immunodeficiencyvirus infection in a cat. J. Am. Vet. Med. Assoc. 194:253–255.
202. Shelton, G. H., C. K. Grant, S. M. Cotter, M. B. Gardner, W. D. Hardy, andR. F. DiGiacomo. 1990. Feline immunodeficiency virus and feline leukemiavirus infections and their relationships to lymphoid malignancies in cats: aretrospective study (1968–1988). J. Acquired Immune Defic. Syndr. 3:623–630.
203. Shelton, G. H., M. L. Linenberger, and J. L. Abkowitz. 1991. Hematologicabnormalities in cats seropositive for feline immunodeficiency virus. J. Am.Vet. Med. Assoc. 199:1353–1357.
204. Shelton, G. H., M. L. Linenberger, C. K. Grant, and J. L. Abkowitz. 1990.Hematologic manifestations of feline immunodeficiency virus infection.Blood 76:1104–1109.
205. Shelton, G. H., R. M. Waltier, S. C. Connor, and C. K. Grant. 1989.Prevalence of feline immunodeficiency virus and feline leukemia virusinfections in pet cats. J. Am. Anim. Hosp. Assoc. 25:7–12.
206. Siebelink, K. H. J., M. L. Bosch, G. F. Rimmelzwaan, R. H. Meloen, andA. D. M. E. Osterhaus. Two different mutations in the envelope protein offeline immunodeficiency virus allow the virus to escape from neutralizationby feline serum antibodies. Vet. Immunol. Immunopathol., in press.
207. Siebelink, K. H. J., I.-H. Chu, G. F. Rimmelzwaan, K. Weijer, A. D. M. E.Osterhaus, and M. L. Bosch. 1992. Isolation and partial characterization ofinfectious molecular clones of feline immunodeficiency virus obtaineddirectly from bone marrow DNA of a naturally infected cat. J. Virol.66:1091–1097.
208. Siebelink, K. H. J., I.-H. Chu, G. F. Rimmelzwaan, K. Weijer, R. vanHerwijnen, P. Knell, H. F. Egberink, M. L. Bosch, and A. D. M. E.Osterhaus. 1990. Feline immunodeficiency virus (FIV) infection in the catas a model for HIV infection in man: FIV-induced impairment of immunefunction. AIDS Res. Hum. Retroviruses 6:1373–1378.
209. Siebelink, K. H. J., J. A. Karlas, G. F. Rimmelzwaan, A. D. M. E.Osterhaus, and M. L. Bosch. A determinant of feline immunodeficiencyvirus involved in CrFK cell tropism. Vet. Immunol. Immunopathol., inpress.
210. Siebelink, K. H. J., G. F. Rimmelzwaan, M. L. Bosch, R. H. Meloen, andA. D. M. E. Osterhaus. 1993. A single amino acid substitution in hyper-variable region 5 of the envelope protein of feline immunodeficiency virusallows escape from virus neutralization. J. Virol. 67:2202–2208.
211. Siebelink, K. H. J., R. W. Windrich, I. Chu, J. Groen, K. Weijer, F. G. C. M.UytdeHaag, and A. D. M. E. Osterhaus. 1990. An enzyme-linked immu-nosorbent assay (ELISA) for the detection of feline immunodeficiency virus(FIV) antigen in cell culture and FIV specific antibodies in feline serum.Dev. Biol. Stand. 72:189–196.
212. Sodora, D. L., E. G. Shpaer, B. E. Kitchell, S. W. Dow, E. A. Hoover, andJ. I. Mullins. 1994. Identification of three feline immunodeficiency virus(FIV) env gene subtypes and comparison of the FIV and human immuno-
deficiency virus type 1 evolutionary patterns. J. Virol. 68:2230–2238.213. Song, W., E. W. Collisson, P. M. Billingsley, and W. C. Brown. 1992.
Induction of feline immunodeficiency virus-specific cytolytic T-cell re-sponses from experimentally infected cats. J. Virol. 66:5409–5417.
214. Sparger, E. E., P. A. Luciw, J. H. Elder, J. K. Yamamoto, L. J. Lowenstine,and N. C. Pedersen. 1989. Feline immunodeficiency virus is a lentivirusassociated with an AIDS-like disease in cats. AIDS 3:S43–S49.
215. Sparger, E. E., B. L. Shacklett, L. Renshaw-Gegg, P. A. Barry, N. C.Pedersen, J. H. Elder, and P. A. Luciw. 1992. Regulation of gene expressiondirected by the long terminal repeat of the feline immunodeficiency virus.Virology 187:165–177.
216. Steinman, R., J. Dombrowski, T. O’Connor, R. C. Montelaro, Q. Tonelli, K.Lawrence, C. Seymour, J. Goodness, N. C. Pedersen, and P. R. Andersen.1990. Biochemical and immunological characterization of the major struc-tural proteins of feline immunodeficiency virus. J. Gen. Virol. 71:701–706.
217. Stephens, E. B., E. J. Butfiloski, and E. Monck. 1992. Analysis of the aminoterminal presequence of feline immunodeficiency virus glycoprotein: effectof deletions on the intracellular transport of gp95. Virology 190:569–578.
218. Stephens, E. B., E. Monck, K. Reppas, and E. J. Butfiloski. 1991. Processingof the glycoprotein of feline immunodeficiency virus: effect of inhibitors ofglycosylation. J. Virol. 65:1114–1123.
219. Sutjipto, S., N. C. Pedersen, C. J. Miller, M. B. Gardner, C. V. Hanson, A.Gettie, M. Jennings, J. Higgins, and P. A. Marx. 1990. Inactivated simianimmunodeficiency virus vaccine fails to protect rhesus macaques fromintravenous or genital mucosal infection but delays the disease course inintravenously exposed animals. J. Virol. 64:2290–2297.
220. Swinney, G. R., J. V. Pauli, B. R. Jones, and C. R. Wilks. 1989. FelineT-lymphotropic virus (FTLV) (feline immunodeficiency virus infection) incats in New Zealand. N. Z. Vet. J. 37:41–43.
221. Talbott, R. L., E. E. Sparger, K. M. Lovelace, W. M. Fitch, N. C. Pedersen,P. A. Luciw, and J. H. Elder. 1989. Nucleotide sequence and genomicorganization of feline immunodeficiency virus. Proc. Natl. Acad. Sci. USA86:5743–5747.
222. Tanabe-Tochikura, A., T. S. Tochikura, J. R. Blakeslee, Jr., R. G. Olsen,and L. E. Mathes. 1992. Anti-human immunodeficiency (HIV) agents arealso potent and selective inhibitors of feline immunodeficiency virus(FIV)-induced cytopathic effect: development of a new method for screen-ing of anti-FIV substances in vitro. Antiviral. Res. 19:161–172.
223. Taniguchi, A., T. Ishida, A. Konno, T. Washizu, and I. Tomoda. 1990.Altered mitogen response of peripheral blood lymphocytes in differentstages of feline immunodeficiency virus infection. Jpn. J. Vet. Sci. 52:513–518.
224. Taniguchi, A., T. Ishida, T. Washizu, and I. Tomoda. 1991. Humoralimmune response to T cell dependent and independent antigens in catsinfected with feline immunodeficiency virus. J. Vet. Med. Sci. 53:333–335.
225. Tenorio, A. P., C. E. Franti, B. R. Madewell, and N. C. Pedersen. 1991.Chronic oral infections of cats and their relationship to persistent oralcarriage of feline calici-, immunodeficiency, or leukemia viruses. Vet.Immunol. Immunopathol. 29:1–14.
226. Thomas, J. B., W. F. Robinson, B. J. Chadwick, I. D. Robertson, and S. A.Beetson. 1993. Association of renal disease indicators with feline immuno-deficiency virus infection. J. Am. Anim. Hosp. Assoc. 29:320–326.
227. Tilton, G. K., T. P. O’Connor, Jr., C. L. Seymour, K. L. Lawrence, N. D.Cohen, P. R. Andersen, and Q. J. Tonelli. 1990. Immunoassay for detectionof feline immunodeficiency virus core antigen. J. Clin. Microbiol. 28:898–904.
228. Tochikura, T. S., K. A. Hayes, C. M. Cheney, A. Tanabe-Tochikura, J. L.Rojko, L. E. Mathes, and R. G. Olsen. 1990. In vitro replication andcytopathogenicity of the feline immunodeficiency virus for feline T4 thymiclymphoma 3201 cells. Virology 179:492–497.
229. Tochikura, T. S., A. Tanabe-Tochikura, K. A. Hayes, A. Lazo, R. T. Bailer,J. R. Blakeslee, Jr., L. J. Lafrado, P. Roy-Burman, R. Pandey, R. G. Olsen,and L. E. Mathes. 1993. Fusion activity dissociated from replication abilityin feline immunodeficiency virus in human cells. J. Acquired Immune Defic.Syndr. 6:1301–1310.
230. Tokunaga, K., Y. Nishino, H. Oikawa, C. Ishihara, T. Mikami, and K.Ikuta. 1992. Altered cell tropism and cytopathicity of feline immunodefi-ciency viruses in two different feline CD4-positive, CD8-negative cell lines.J. Virol. 66:3893–3898.
231. Tomonaga, K., T. Miyazawa, J.-I. Sakuragi, T. Mori, A. Adachi, and T.Mikami. 1993. The feline immunodeficiency virus ORF-A gene facilitatesefficient viral replication in established T-cell lines and peripheral bloodlymphocytes. J. Virol. 67:5889–5895.
232. Tomonaga, K., J. Norimine, Y.-S. Shin, M. Fukasawa, T. Miyazawa, A.Adachi, T. Toyosaki, T. Kawaguchi, C. Kai, and T. Mikami. 1992. Identi-fication of a feline immunodeficiency virus gene which is essential forcell-free virus infectivity. J. Virol. 66:6181–6185.
233. Tomonaga, K., Y.-S. Shin, M. Fukasawa, T. Miyazawa, A. Adachi, and T.Mikami. 1993. Feline immunodeficiency virus gene expression: analysis ofthe RNA splicing pattern and the monocistronic rev mRNA. J. Gen. Virol.74:2409–2417.
234. Tompkins, M. B., P. D. Nelson, R. V. English, and C. Novotney. 1991. Early
events in the immunopathogenesis of feline retrovirus infections. J. Am.Vet. Med. Assoc. 199:1311–1315.
235. Torten, M., M. Franchini, J. E. Barlough, J. W. George, E. Mozes, H. Lutz,and N. C. Pedersen. 1991. Progressive immune dysfunction in cats experi-mentally infected with feline immunodeficiency virus. J. Virol. 65:2225–2230.
236. Toyosaki, T., T. Miyazawa, T. Furuya, K. Tomonaga, Y.-S. Shin, M. Okita,Y. Kawaguchi, C. Kai, S. Mori, and T. Mikami. 1993. Localization of theviral antigen of feline immunodeficiency virus in the lymph nodes of cats atthe early stage of infection. Arch. Virol. 131:335–347.
237. Tozzini, F., D. Matteucci, P. Bandecchi, F. Baldinotti, A. Poli, M. Pistello,K. H. J. Siebelink, L. Ceccherini-Nelli, and M. Bendinelli. 1992. Simple invitro methods for titrating feline immunodeficiency virus (FIV) and FIVneutralizing antibodies. J. Virol. Methods 37:241–252.
238. Tozzini, F., D. Matteucci, P. Bandecchi, F. Baldinotti, K. H. J. Siebelink, A.Osterhaus, and M. Bendinelli. 1993. Neutralizing antibodies in cats in-fected with feline immunodeficiency virus. J. Clin. Microbiol. 31:1626–1629.
239. Tsoukas, C. M., and N. F. Bernard. 1993. Markers predicting progression ofhuman immunodeficiency virus-related disease. Clin. Microbiol. Rev. 7:14–28.
240. Ueland, K., and L. L. Nesse. 1992. No evidence of vertical transmission ofnaturally acquired feline immunodeficiency virus infection. Vet. Immunol.Immunopathol. 33:301–308.
241. Verschoor, E., A. van Vliet, H. F. Egberink, W. Hesselink, M. C. Horzinek,and A. de Ronde. 1993. Expression of feline immunodeficiency virus gag andenv precursor proteins in Spodoptera frugiperda cells and their use inimmunodiagnosis. J. Clin. Microbiol. 31:2350–2355.
242. Verschoor, E. J., A. L. W. van Vliet, H. F. Egberink, W. Hesselink, W. E. vanAlpen, I. Joosten, J. C. P. Boog, M. C. Horzinek, and A. de Ronde.Vaccination against feline immunodeficiency virus using fixed infected cells.Vet. Immunol. Immunopathol., in press.
243. von Kolbl, S., and W. Schuller. 1989. Serologische Untersuchungen zumVorkommen des felin Immundefizienzvirus (FIV) bei Katzen in Osterreich.Wien. Tieraerztl. Mschr. 76:185–189.
244. Wagaman, P. C., C. S. Hasselkus-Light, M. Henson, D. L. Lerner, T. R.Phillips, and J. H. Elder. 1993. Molecular cloning and characterization ofdeoxyuridine triphosphatase from feline immunodeficiency virus (FIV).Virology 196:451–457.
245. Wasmoen, T., S. Armiger-Luhman, C. Egan, V. Hall, H.-J. Chu, L. Chavez,and W. Acree. 1992. Transmission of feline immunodeficiency virus frominfected queens to kittens. Vet. Immunol. Immunopathol. 35:83–93.
246. Waters, L., C. D. Hopper, T. J. Gruffydd-Jones, and D. A. Harbour. 1993.Chronic gingivitis in a colony of cats infected with feline immunodeficiencyvirus and feline calicivirus. Vet. Rec. 132:340–342.
247. Weiss, R. A. 1993. How does HIV cause AIDS? Science 260:1273–1279.248. Wheeler, D. W., T. W. Mitchell, P. W. Gasper, M. C. Barr, and L. R.
Whalen. 1991. FIV infection associated with neurologic abnormalities, p.31. In Abstracts of the 1st International Conference of Feline Immunode-ficiency Virus Researchers 1991. University of California, Davis.
249. Willett, B. J., M. J. Hosie, J. J. Callanan, J. C. Neil, and O. Jarrett. 1993.
Infection with feline immunodeficiency virus is followed by the rapidexpansion of a CD81 lymphocyte subset. Immunology 78:1–6.
250. Willett, B. J., M. J. Hosie, T. H. Dunsford, J. C. Neil, and O. Jarrett. 1991.Productive infection of T-helper lymphocytes with feline immunodeficiencyvirus is accompanied by reduced expression of CD4. AIDS 5:1469–1475.
251. Willett, B. J., M. J. Hosie, O. Jarrett, and J. C. Neil. 1994. Identification ofa putative cellular receptor for feline immunodeficiency virus as the felinehomologue of CD9. Immunology 81:228–233.
252. Willett, B. J., and O. Jarrett (ed.). Feline immunology and immunodefi-ciency, Oxford University Press, in press.
253. Witt, C. J., T. R. Moench, A. M. Gittelsohn, B. D. Bishop, and J. E. Childs.1989. Epidemiologic observations on feline immunodeficiency virus andToxoplasma gondii coinfection in cats in Baltimore, Md. J. Am. Vet. Med.Assoc. 194:229–233.
254. Yamamoto, J. K., C. D. Ackley, H. Zochlinski, H. Louie, E. Pembroke, M.Torten, H. Hansen, R. Munn, and T. Okuda. 1991. Development ofIL-2-independent feline lymphoid cell lines chronically infected with felineimmunodeficiency virus: importance for diagnostic reagents and vaccines.Intervirology 32:361–375.
255. Yamamoto, J. K., H. Hansen, E. W. Ho, T. Y. Morishita, T. Okuda, T. R.Sawa, R. M. Nakamura, and N. C. Pedersen. 1989. Epidemiologic andclinical aspects of feline immunodeficiency virus infection in cats from thecontinental United States and Canada and possible mode of transmission.J. Am. Vet. Med. Assoc. 194:213–220.
256. Yamamoto, J. K., T. Hohdatsu, R. A. Holmsted, R. Pu, H. Louie, H. A.Zochlinski, V. Acevedo, H. M. Johnson, G. A. Soulds, and M. B. Gardner.1993. Experimental vaccine protection against homologous and heterolo-gous strains of feline immunodeficiency virus. J. Virol. 67:601–605.
257. Yamamoto, J. K., T. Okuda, C. D. Ackley, H. Louie, E. Pembroke, H.Zochlinski, R. J. Munn, and M. B. Gardner. 1991. Experimental vaccineprotection against feline immunodeficiency virus. AIDS Res. Hum. Retro-viruses 7:911–922.
258. Yamamoto, J. K., N. C. Pedersen, E. W. Ho, T. Okuda, and G. H. Theilen.1988. Feline immunodeficiency syndrome—a comparison between felineT-lymphotropic lentivirus and feline leukemia virus. Leukemia 2:204S–215S.
259. Yamamoto, J. K., E. Sparger, E. W. Ho, P. R. Andersen, T. P. O’Connor,C. P. Mandell, L. Lowenstine, R. Munn, and N. C. Pedersen. 1988.Pathogenesis of experimentally induced feline immunodeficiency virusinfection in cats. Am. J. Vet. Res. 49:1246–1258.
260. Yates, K. M., L. J. Rosenberg, C. K. Harris, D. C. Bronstad, G. K. King,G. A. Biehle, B. Walker, C. R. Ford, J. E. Hall, and I. R. Tizard. 1992. Pilotstudy of the effect of acemannan in cats infected with feline immunodefi-ciency virus. Vet. Immunol. Immunopathol. 35:177–189.
261. Zenger, E. 1990. Clinical findings in cats with feline immunodeficiencyvirus. Feline Pract. 18:25–28.
262. Zenger, E., W. C. Brown, W. Song, A. M. Wolf, N. C. Pedersen, M.Longnecker, J. Li, and E. W. Collisson. 1993. Evaluation of cofactor effectof feline syncytium-forming virus in feline immunodeficiency virus infec-tion. Am. J. Vet. Res. 56:713–718.
