Dose response studies of acute feline immunodeficiency virus PPR strain infection in cats

Veterinary Microbiology 76 (2000) 311±327

Dose response studies of acute feline immunode®ciency virus PPR strain infection in cats Regina M. Hokansona, Julie TerWeeb, In-Soo Choia, Joan Coatesc, Hansi Deand, D.N. Reddye, Alice M. Wolfc, Ellen W. Collissona,* a

Department of Veterinary Pathobiology, The Texas Veterinary Medical Center, College of Veterinary Medicine, Texas A&M University, College Station, TX 77843-4467, USA b Biocore, 5551 Colby, Lincoln, NE 60504, USA c Department of Small Animal Medicine and Surgery, College of Veterinary Medicine, Texas A&M University, College Station, TX 77843-4467, USA d Endorex Corporation, 28101 Ballard Drive, Lake Forest, IL 60045, USA e Schering-Plough Animal Health, Elkhorn, NE 60504, USA Received 13 October 1999; received in revised form 2 May 2000; accepted 19 June 2000

Abstract The effects of virus dose on host response were evaluated for the PPR strain of feline immunode®ciency virus (FIV-PPR). Speci®c pathogen-free cats were inoculated intravenously with 50, 250 or 1250 TCID50 of FIV-PPR. Two weeks after inoculation, virus was detected in 106 peripheral blood mononuclear cells (PBMCs) of all infected animals, and the CD4‡:CD8‡ T lymphocyte ratios fell from greater than 2 to approximately 1 in all infected animals within the ®rst 8 weeks after infection. Provirus detected in all groups using PCR and 103 PBMC was biphasic. Nine of 15 animals were positive between weeks 2 and 4 p.i. and 14 of 15 were positive by week 8 p.i. Transient lymphadenopathy was detected in most cats receiving 1250 TCID50 and the 250 TCID50 of virus, whereas no lymphadenopathy was detected in the 50 TCID50 group or the ®ve uninfected cats. Animals that had received the largest dose seroconverted earliest (on average at week 4.0) and those receiving the least seroconverted last (on average at week 5.6). Neither neutropenia nor lymphopenia were detected. FIV-speci®c CTL responses of memory effector cells could be detected in animals receiving all three doses but was highly variable among individual animals. Neurological manifestations determined after 15 weeks p.i. were observed in most infected cats, including two of the three that had received 50 TCID50 of virus. However, the observed neurologic abnormalities were markedly less severe in the animals receiving the least amount of *

Corresponding author. Tel.: ‡1-409-845-4122; fax: ‡1-409-862-1088. E-mail address: [email protected] (E.W. Collisson). 0378-1135/00/$ ± see front matter # 2000 Elsevier Science B.V. All rights reserved. PII: S 0 3 7 8 - 1 1 3 5 ( 0 0 ) 0 0 2 6 3 - 7

312

R.M. Hokanson et al. / Veterinary Microbiology 76 (2000) 311±327

virus. Therefore, lymphadenopathy and neurologic signs of illness were less severe and seroconversion was slower in the animals that received the lowest dose compared with those receiving the 250 and 1250 TCID50 doses of the FIV-PPR strain. # 2000 Elsevier Science B.V. All rights reserved. Keywords: Feline immunode®ciency virus; PPR strain; Cat; CD4:CD8 ratios; Neurologic signs of illness; Cytotoxic T lymphocytes

1. Introduction Feline immunode®ciency virus (FIV), a lymphotropic retrovirus, is a member of the lentivirus subfamily of retroviruses. FIV has been shown to cause an immunode®ciency disease in cats similar to human AIDS also caused by a lentivirus, human immunode®ciency virus (HIV) ÿ1 (Pedersen et al., 1987; Yamamoto et al., 1988; Matsumara et al., 1993). Serological surveys have indicated that FIV infection occurs in felids world-wide, including domestic cats, pumas of North America, and the large cats of Africa (Carpenter et al., 1996; Osofsky et al., 1996; Pedersen and Barlough, 1991). FIV-speci®c antibodies have been identi®ed in archived sera from the late 1960s and early 1970s (Pedersen et al., 1987). Due to many similarities between HIV and FIV, FIV research is relevant in the study of lentivirus pathogenesis and for vaccine studies (Bendinelli et al., 1995; Vitkovic et al., 1995; Jarrett, 1996). The PPR strain (FIV-PPR) was isolated from a cat in the USA that exhibited AIDS-like signs in the absence of a feline leukemia virus infection (Philips et al., 1990). This strain has been found to have distinct in vitro host cell preference compared to the ®rst USA isolate, Petaluma (Philips et al., 1990). FIV-PPR readily infects primary feline peripheral blood mononuclear cells (PBMCs), but unlike Petaluma, does not directly infect Crandell feline kidney (CrFK) or astroglia cells (Philips et al., 1990). This difference correlated with the ability for Petaluma envelope proteins to induce syncytium formation in CrFK cells (Philips et al., 1990; Poss et al., 1992; Pancino et al., 1995). In contrast, both strains infect CD4ÿCD8‡ and CD4‡CD8ÿ T lymphocytes (Brown et al., 1991). Some strains of FIV, such as FIV-PPR, are neuropathic (Dow et al., 1990; Philips et al., 1994, 1996; Tompkins et al., 1991). Behavioral changes have been observed in FIVinfected cats (Philips et al., 1994, 1996; Prospero-Garcia et al., 1994; Vitkovic et al., 1995), as well as motor neuronal dysfunction (Dow et al., 1990; Philips et al., 1996). Infection has been produced in cats following inoculation of infectious FIV-PPR cDNA provirus (Sparger et al., 1994, 1997). Varying concentrations of inoculated cDNA provirus resulted in dose-responsive kinetics of viral isolation and seroconversion (Sparger et al., 1997). Whereas all cats inoculated with 100 or 300 mg of cDNA became infected, the numbers positively infected with the lowest dose, 30 mg was incomplete. The dose-response effects of whole virus acute infection have not been reported for FIV-PPR. In this study, cats were inoculated with varying doses of whole infectious FIVPPR strain virus. The clinical signs of illness, presence of provirus, and immunologic responses were examined for 8 weeks post-inoculation (p.i.). Cytotoxic T lymphocyte assays and neurological exams were performed at 15 weeks p.i. and later.

R.M. Hokanson et al. / Veterinary Microbiology 76 (2000) 311±327

313

2. Materials and methods 2.1. Production of FIV-PPR DNA from a molecular clone Standard procedures were used to grow Eschericia coli which contained an FIV-PPR clone (NIH AIDS Reference Reagent Collection, McKesson BioServices, Rockville, MD). Brie¯y, as described for the QIAgen miniprep spin kit (Santa Claret, CA), the culture was centrifuged to pellet the bacteria and the plasmid DNA was collected using a modi®ed alkaline lysis procedure. The DNA concentration of varying dilutions was determined with a spectrophotometer at OD260/280. 2.2. Production of infectious FIV-PPR virus To facilitate the production of infectious virus from the FIV-PPR clone, the plasmid was transfected into CrFK cells (John Elder, Scripps Research Institute, La Jolla, CA, personal communication). Since FIV-PPR does not naturally infect CrFK cells (Philips et al., 1990), primary feline lymphocytes were added after 24 h to amplify the virus. In more detail, following instructions accompanying the Lipofectamine transfection system (GIBCO-BRL, Gaithersburg, MD), 2 mg of the FIV-PPR DNA was combined with 25 ml of Lipofectamine in Opti-MEM serum-free media (GIBCO-BRL, Gaithersburg, MD) without antibiotics. The newly formed liposomes were then added to a 60% con¯uent layer of CrFK cells propagated in a 25 cm2 tissue culture ¯ask. After 3 h incubation at 378C with 5% CO2, normal CrFK culture medium containing twice the normal amount of fetal bovine serum was added to the cells. The following day, normal, uninfected feline PBMC were added to the CrFK cells and co-cultured for 24 h. The PBMC were removed from co-culture and grown for 10 days in complete RPMI 1640 plus 100U recombinant human interleukin 2 (IL-2) per ml (Becton-Dickinson, Bedford, MA). Collected supernatants were centrifuged, ®ltered and stored at ÿ708C. 2.3. Viral titration Normal PBMC were collected and isolated as described below. After Concanavalin A (Con A) stimulation, the cells were washed and 4  106 cells/ml, 250 ml/well, were added in quadruplicate to 48 well tissue culture plates. Supernatants from PBMC infected with virus were diluted between 103- and 108-fold in complete RPMI before adding 250 ml to each well. On days 3 and 8, an additional 250 ml of complete media and IL-2 were added. On day 10, 200 ml of supernatant from each well was tested for virus production using PetCheck capsid antigen ELISA kit (IDEXX, Portland, ME). Virus titer, calculated using the Reed and Muench calculation of 50% endpoints, was 106.3 (Burleson et al., 1992). 2.4. Animals and virus inoculations Twenty speci®c pathogen-free (SPF), 13-week-old male kittens were purchased from Harlan-Sprague Dawley (Indianapolis, IN). They were housed at Texas A&M University's Laboratory Animal Research and Resource support facilities in temperature

314

R.M. Hokanson et al. / Veterinary Microbiology 76 (2000) 311±327

controlled SPF conditions for the duration of the experiment. The kittens were fed a feline growth formula diet and allowed water ad libitum. The kittens were randomly divided into four groups of ®ve individuals 1 week before inoculation to re-establish dominance hierarchies. The negative control group was inoculated i.v. into the right cephalic vein with buffered saline only. The three treatment groups received 1 ml of buffered saline containing the PPR strain of FIV at a titer of 50, 250, or 1250 TCID50 into the right cephalic vein. For ease of blood collection, the negative control and 50 TCID50 groups were inoculated on the ®rst day and the 250 and 1250 TCID50 groups were inoculated 1 day later. Freshly thawed virus was used with each group. Virus samples from each vial used, along with the dilutions used, were again titered on T lymphocytes beginning the day of the inoculations. Nine weeks after inoculation, the cats in the 250 TCID50 group and two in each of the other inoculation groups were euthanized. Three uninfected cats were adopted out at 9 weeks and two at 16 weeks p.i. with saline. At 16 weeks p.i., the remaining cats in the 50 TCID50 group and a third cat in the 1250 TCID50 group were euthanized. In a pilot experiment, two cats, A306 and A308, had been similarly inoculated 4 months earlier with 250 TCID50 of the PPR strain of FIV. Sample collections for these animals were more limited compared with the primary study described in this paper. 2.5. Physical and neurological examinations Complete physical exams were performed upon arrival, at day ÿ1 and day 57 on all animals in the study. Following inoculation, body temperature and general health were monitored twice a day for 3 weeks. The body temperatures were then recorded once per day for the next 3 weeks and once per day every other day for the ®nal 2 weeks. Neurologic dysfunction in infected cats was assessed subjectively by the clinical neurologic examination (Oliver et al., 1997). Neurologic examinations included observation (mentation, posture, gait); cranial nerve evaluation; postural reaction testing (conscious proprioceptive positioning and hopping, wheelbarrowing, and extensor postural thrust reactions); assessment of spinal re¯exes, i.e. myotatic, ¯exor withdrawal; and sensory evaluation. Myotatic refers to a tendon tap test or stretching of the muscle spindles. Gait evaluation was assessed as normal, ataxic or paretic. Postural reactions were assessed as normal, decreased (mild, moderate, and severe) or absent. Spinal re¯exes were assessed as normal, decreased or absent. 2.6. Anti-FIV capsid antibody assay Complete blood counts were performed for every 2 weeks by the Texas A&M University, College of Veterinary Medicine's Clinical Pathology Laboratory in order to monitor lymphocyte and neutrophil levels. The Duncan new multiple range test was used to determine signi®cant differences (SuperANOVA version 1.11 by Abacus Concepts of Berkeley, CA). Detection of antibodies against FIV-PPR in plasma from each weekly blood collection was evaluated using the IDEXX PetCheck anti-capsid antibody ELISA kit (Portland, ME).

R.M. Hokanson et al. / Veterinary Microbiology 76 (2000) 311±327

315

2.7. Preparation of PBMCs Whole blood was drawn by jugular venipuncture once per week for 2 weeks prior to and on the day of inoculation. The pre-inoculation PBMC were separated from the whole blood by gradient centrifugation using Histopaque 1.077 (Sigma, St. Louis, MO) and washed twice with cold Alsever's solution containing 1% penicillin±streptomycin. PBMC collected during weeks 2, 4, 6, and 8 and without Con A stimulation were used to determine CD4‡:CD8‡ ratios, proviral DNA PCR, or complete blood counts. 2.8. CD4‡:CD8‡ lymphocyte ratios Individual CD4‡ and CD8‡ T-lymphocyte ratios were determined at weeks 0, 2, 4, 6, and 8 (Tompkins et al., 1991; Song et al., 1992; Zenger et al., 1993). The freshly isolated T-lymphocytes (1  106 ) were washed twice in cold PBS-0.1% sodium azide and divided among ®ve 12 mm  75 mm polystyrene tubes. Twelve and a half ml of 2 mg/ml goat IgG (Sigma, St. Louis, MO) were added to block Fc receptors and reduce nonspeci®c binding of the labeled antibodies. The cells were incubated for 10 min on ice. A 1:25 dilution of mouse anti-cat CD4, CD8 or CD5 monoclonal antibodies (Southern Biotechnology, Birmingham, AL) in cold PBS-0.1% sodium azide was added to respective tubes for a ®nal dilution of 1:50. After incubating on ice for 20 min, the cells were washed with 3 ml of cold PBS-0.1% sodium azide and centrifuged at 1600g for 3 min in a low speed centrifuge. Fluorescein labeled goat anti-mouse IgG (H ‡ L) (KPL, Gaithersburg, MD) in cold PBS-0.1% sodium azide was added to a ®nal concentration of 1:50. The cells were again incubated on ice for 20 min and washed with PBS-0.1% sodium azide. After adding 300 ml of 3.7% formaldehyde to each tube, the rack of tubes was wrapped in plastic ®lm and aluminum foil, and kept at 48C until the samples were examined the next day with a FACscalibur ¯ow cytometer (Becton-Dickinson, Bedford, MA) at the Department of Veterinary Pathobiology core facility, Texas A&M University. 2.9. Virus detection by isolation and proviral PCR ampli®cation Attempts to isolate virus from fresh PBMC were made each week from week 0 (preinoculation) to week 8 p.i. Unstimulated cells from each cat were co-cultured with Con A (Sigma, St. Louis, MO) stimulated cells that had been collected from uninoculated littermates. The co-cultured cells were passaged each week, and supernatants were collected for 4 weeks and stored at ÿ808C. PetCheck FIV antigen detection ELISA kits (IDEXX, Portland, ME) were used to detect FIV capsid proteins in the supernatants. In order to amplify the proviral DNA, 1  103 unstimulated PBMC were washed twice in PBS and digested using sodium dodecyl sulfate, Triton X-100 and 15 ml of 20 mg proteinase K/ml (Roche Molecular Biochemicals, Indianapolis, IN) for 30 min at 378C. The proteinase K was heat inactivated at 948C for 15 min (Li et al., 1995). Thirty microliters of the cell digests were combined with 18 ml of PCR stock buffer (25 mM MgCl2, 10 mM each of ATP, GTP, CTP and TTP plus 10X thermophilic buffer) and 2.5 units of Taq DNA polymerase (all components purchased from Promega, Madison, WI). The primers were used to amplify a 508 base pair (b.p.) region in the capsid region. Sixty cycles of ampli®cation were performed

316

R.M. Hokanson et al. / Veterinary Microbiology 76 (2000) 311±327

in two (30 cycle) rounds (Li et al., 1995). Additional primer, buffer, and Taq DNA polymerase were added to 20 ml of samples for each session. To increase speci®city and sensitivity, a third 30 cycle round was performed using a primer pair which ampli®ed a 400 b.p. region within the original 508 b.p. region (Li et al., 1995). The products were examined on the 1% agarose gel after the third round of ampli®cation. Controls with cells from uninfected cats and from known infected cats were always included. Often both PCR products (400 and 508 b.p.) could be seen after the third round of PCR. 2.10. Cytotoxic T lymphocyte (CTL) assays Effector cell CTL activity was determined after co-culturing for 2 weeks with stimulator cells (Song et al., 1992; Li et al., 1995). T-lymphocytes from PBMC were gradient puri®ed, stimulated 72 h with 5 mg/ml Con A and cultured in complete RPMI 1640 with 100U human recombinant (h.r.) IL-2 per ml (Collaborative Biosciences, Bedford, MA). Tissue culture supernatant of 2±500 ml containing FIV-PPR (OD630 value of 3.5‡) were added to the PBMC. Active infection was con®rmed on day 3 p.i. by Petcheck FIV capsid antigen ELISA kit (IDEXX, Portland, ME). One-third of the cells were used, the next day, as stimulator cells. The remaining two-thirds were cryopreserved in 10% DMSO ‡ 90% FBS. Three or four days before, the cryopreserved infected cells were to be used as stimulator or target cells, they were thawed quickly and cultured with h.r. IL-2. Effector cells were cultured for 14 days with irradiated (10 000 rad) autologous PPR infected cells (Song et al., 1992; Li et al., 1995), without Con A. Before determining the CTL activity, the target cells were labeled with 5 mCi of 111indium oxine (MediphysicsAmersham, San Antonio, TX) for one and half hour at 378C. The effector and target cells were combined in triplicate in ratios varying from 5:1 to 100:1 effector to target cells in 200 ml complete media in V-bottomed 96 well tissue culture plates (Corning-Costar, Rochester, NY). Uninfected cells from each cat were used as negative controls. FIV-PPR infected target cells from a second cat were included as heterologous (mismatched) controls. After gentle (150g), 4 min of centrifugation, the cells were incubated for 4 h at 378C with 5% CO2. The plates were then centrifuged at 450g for 10 min to pellet the cells. One hundred microliter of supernatant from each well were counted with a Packard Cobra Autogamma gamma counter (Meridian, CT). Maximum release consisted of radiolabeled target cells in 200 ml media lysed with 3% triton X-100. Spontaneous release (background) was determined with radiolabeled target cells in 200 ml media after incubation at 378C, 5% CO2 for 4 h. Results are presented as percent killing and calculated as [(sampleÿspontaneous release)/(maximum releaseÿspontaneous release)]100%. PBMC and CTL from cat A306 were included as a positive control for the assay. This animal reproducibly had detectable CTL activity speci®c for FIV (Choi et al., 2000; personal observations). 3. Results Cats were inoculated i.v. with 50, 250, or 1250 TCID50 of the FIV-PPR strain. Numerous parameters, including clinical illness, were evaluated p.i. Two animals in a

R.M. Hokanson et al. / Veterinary Microbiology 76 (2000) 311±327

317

Fig. 1. The average CD4‡:CD8‡ ratio of T lymphocytes collected from PBMC of each inoculation group.

pilot study were inoculated with the 250 TCID50 dose of the FIV-PPR strain. The parameters examined and sample collections for the pilot study were limited compared with the larger study. 3.1. CD4‡:CD8‡ T lymphocyte ratios and complete blood cell evaluations CD4‡:CD8‡ T lymphocyte ratios in PBMC were determined prior to inoculation of virus and every 2 weeks p.i. (Fig. 1). The decrease in and inversion of CD4‡:CD8‡ ratios has often been observed in the asymptomatic phase of FIV infection (Barlough et al., 1991). Whereas decreases could be observed in the CD4‡:CD8‡ T lymphocyte ratios of the infected cats through 8 weeks p.i. and not in the uninfected controls, there were no differences between the inoculated groups. Complete blood counts indicated that the concentrations of lymphocytes (<3500 lymphocytes/ml) and neutrophils (<2500 neutrophils/ml) observed in these animals did not ¯uctuate from normal values. 3.2. Physical and neurological examinations Physical examinations were normal and body temperatures remained below 102.58C in all groups throughout the study except for cats within the 50 TCID50 inoculation group on day 12 p.i. During the afternoon of day 12 p.i., the temperatures ranged from 102.7 to 103.98C. The observed behavior of all cats was normal, i.e. there was no change in appetite, playing activity and grooming. A single cat in the 50 TCID50 group, E210, had sporadic temperatures above 102.58C between days 5 and 7, again without changes in behavior. Lymphadenopathy in super®cial lymph nodes was observed in three of ®ve cats receiving 1250 TCID50 of virus and two of ®ve cats receiving 250 TCID50 of virus, but not in cats receiving the lowest dose of PPR virus or the controls (Table 1). The lymphadenopathy was detectable 4 and 5 days p.i. and on day 14 p.i. for one animal receiving the 250 TCID50 dose. Another cat in the 250 TCID50 group experienced mild upper respiratory problems from day 7 to 10 p.i., but without palpable lymph node enlargement. Similarly enlargements of lymph nodes of cats A306 and A308 (inoculated with 250 TCID50) were observed during acute infection.

318

R.M. Hokanson et al. / Veterinary Microbiology 76 (2000) 311±327

Table 1 Palpable lymph nodes after inoculation with FIV-PPR virus Cat

Dose

Day p.i.

Lymphadenopathy

E244 E254 E260 E266 E300 E210 E254 E242 E290 E296 E206 E258 E272 E264 E298 E238 E268 E274 E284 E288

None None None None None 50 50 50 50 50 250 250 250 250 250 1250 1250 1250 1250 1250

± ± ± ± ± ± ± ± ± ± 4 ± ± ± 4 ± ± 4 4 4

± ± ± ± ± ± ± ± ± ± Subscapular and popliteal ± ± ± Subscapular ± ± Subscapular and submandibular Subscapular and popliteal Subscapular

and 5

and 14 and 5 and 5 and 5

Neurologic examinations were performed for the normal control (E244, E266, and E260) and infected (50 TCID50 and 1250 TCID50 groups) cats that were not euthanized at week 9 p.i. (Table 2). Three cats (E238, E284, and E274) in the 1250 TCID50 group and three cats (E210, E290, and E254) in the 50 TCID50 group were evaluated at 15 weeks p.i. The neurological examination was repeated at 29 weeks p.i. for cats E238 and E284. The two infected cats A306, A308, receiving 250 TCID50 of PPR virus 4 months prior to the other animals, were evaluated at 25, 35 and 46 weeks p.i. Neurologic examinations revealed pelvic limb dysfunction in two of three cats (E210, E254) in the 50 TCID50 group, both cats receiving 250 TCID50, and two of three in the 1250 TCID50 group 15 weeks after inoculation. Neurologic dysfunction was observed to be transient upon reevaluation of two cats (E238, E284) in the 1250 TCID50 group 29 weeks after inoculation. Persistent behavioral abnormalities consisting of fearful aggression and agitation were observed in E238 (inoculated with 1250 TCID50 of virus). Therefore, dif®cult to statistically evaluate, the amount and severity of neurologic problems were more obvious in the animals receiving the 250 and 1250 TCID50 doses of virus. Gait, spinal re¯exes and pain perception were normal in all the inoculated animals. All parameters were normal in the uninoculated controls. 3.3. Seroconversion The control animals remained seronegative for FIV throughout the course of the experiment (Fig. 2). The week that antibodies to the capsid protein were ®rst identi®ed

Table 2 Neurological evaluations of infected cats Mental status

Gait

Postural reactionsa

Cranial nerve evaluation

50 TCID50 inoculated cats E210, 15 E290, 15 E254, 15

Alert Alert Alert

Normal Normal Normal

Normal Normal

1250 TCID50 inoculated cats E238, 15

EPT: mild decrease Normal Hopping: mild decrease R PL; EPT: mild decrease R PL

Alert, agitated

Spastic PL

Normal

E238, 29

Alert, very agitated

Normal

E284, 15

Alert

Normal

E284, 29 E274, 15

Alert Alert

Normal Normal

CP: mild decrease PL; hopping: decreased PL R > L; EPT: moderate decrease CP: mild decrease TL and PL; hopping: mild decrease CP: mild decrease R PL; hopping: mild decrease R TL and PL; EPT: mild decreased PL Normal Normal

250 TCID50 from an earlier experiment A306, 25 Alert

Normal

A306, 35

Alert

Normal

A306, 46

Alert, fearful

Normal

A308, 25 A308, 35

Alert Alert

Normal Normal

A308, 46

Alert

Normal

a

CP: normal; hopping: mild decrease PL; EPT: mild decrease CP: normal; hopping: mild decrease PL; EPT: mild decrease CP: mild decrease R PL; hopping: mild decrease PL; EPT: mild decrease Normal; EPT: mild decrease CP: mild decrease PL; hopping: mild decrease PL; EPT: moderate decrease Normal

Palpebral fissure narrow OD Normal Normal Normal Normal Normal Normal Normal Normal

R.M. Hokanson et al. / Veterinary Microbiology 76 (2000) 311±327

Cat, week post inoculation

Normal

CP: conscious proprioceptive positioning, EPT: extensor postural thrust, PL: pelvic limb, TL: thoracic limb, R: right, L: left. Uninfected cats were normal. 319

320

R.M. Hokanson et al. / Veterinary Microbiology 76 (2000) 311±327

Fig. 2. Initial times p.i. that antibodies speci®c for the FIV capsid protein were detected.

varied with each group. The average time of seroconversion was 4.0 weeks for the cats in the 1250 TCID50 group, 4.6 weeks for the cats in the 250 TCID50 group, and 5.6 weeks for the cats is the 50 TCID50 group. Data were analyzed using one-way analysis of variance using the Duncan new multiple range analysis for comparing treatment means. A signi®cant difference with P > 0:05 was determined between the group of cats given 50 TCID50 and those given 250 TCID50 or 1250 TCID50 of virus. The time of seroconversion or detection of FIV-speci®c antibody was dose dependent, although not statistically signi®cant between the two groups given the highest amount of virus (Fig. 2). Antibody detected was speci®c for the capsid protein (IDEXX, Portland, ME). Neither antibody speci®c for other FIV proteins nor speci®c immunoglobulins were determined. Both A306 and A308 had detectable antibody for FIV by week 6 p.i. 3.4. Virus isolation and PCR detected proviral By day 7 of culture, two of ®ve samples taken on day 7 p.i. from the animals receiving the lowest dose were positive for the presence of FIV capsid by the antigen capture ELISA (IDEXX, Portland, ME), three of ®ve samples taken on day 7 p.i. from the animals receiving the middle dose were positive, and four of ®ve samples taken on day 7 p.i. from the animals receiving the highest dose were positive. Virus was detectable from day 7 p.i. in samples from all inoculated animals after 14 days of culture. Subsequent samples drawn for isolation were positive after 7 days of culture for the remainder of the experiment. There were no discernable differences among the treatment groups in the success of isolating virus from the 5  106 cell samples collected. Although weekly isolations attempts were not done, both A306 and A308 had detectable virus by at least 6 weeks p.i. PCR was used to detect FIV provirus from weekly collections of PBMC consisting of only 103 cells. Therefore, the sample cell numbers were 5000-fold less than samples used for isolation of virus. Provirus was detected in 103 PBMC from all ®ve of the animals receiving 250 TCID50 and 1250 TCID50 doses (Table 3). PBMC from the one member of the 250 TCID50 group did not have detectable provirus until 8 weeks p.i. Four of ®ve

R.M. Hokanson et al. / Veterinary Microbiology 76 (2000) 311±327

321

Table 3 PCR results of provirus extracted from 103 PBMC TCID50 of PPR inoculated

Cat ID

No virus control

Week post-infection 0

1

2

3

4

5

6

7

8

E244 E252 E260 E266 E300

±a ± ± ± ±

± ± ± ± ±

± ± ± ± ±

± ± ± ± ±

± ± ± ± ±

± ± ± ± ±

± ± ± ± ±

± ± ± ± ±

± ± ± ± ±

50

E210 E242 E254 E290 E296

± ± ± ± ±

± ± ± ± ±

± ± ± Pos. Pos.

± ± ± ± ±

± ± Pos. ± ±

± ± ± ± ±

± ± ± Pos. ±

Pos.b ± Pos. Pos. Pos.

Pos. ± Pos. Pos. Pos.

250

E206 E258 E264 E276 E298

± ± ± ± ±

± ± ± ± ±

± ± ± Pos. Pos.

± ± ± Pos. Pos.

± ± ± ± ±

± ± ± ± ±

± ± ± Pos. Pos.

Pos. Pos. ± Pos. Pos.

Pos. Pos. Pos. Pos. Pos.

1250

E238 E268 E274 E284 E288

± ± ± ± ±

± ± ± ± ±

Pos. Pos. Pos. ± ±

± ± ± Pos. ±

± ± ± ± ±

± ± ± ± ±

Pos. Pos. ± Pos. Pos.

± Pos. Pos. Pos. ±

Pos. Pos. Pos. Pos. Pos.

a b

Represents negative detection of PCA products. Represents sera in which antibody for the capsid protein was detected by ELISA.

individuals in the 50 TCID50 group had detectable proviral DNA. With the number of cells examined, all animals were identi®ed as positive by PCR from weeks 2 to 4, but were negative during weeks 5, 6, or 5 and 6. By week 8, 14 of 15 animals again had enough provirus to be detected by this method. By week 8 p.i., 14 of 15 infected animals again had enough provirus to be detected by this method. Proviral DNA was not detected in lymphocytes from any of the uninfected controls. The sporadic nature of provirus detection could be due to early transient control of infection and the small number of lymphocytes used (103 lymphocytes) compared with the 5  106 lymphocytes used for virus isolation. No attempts were made to quantify the viral load. 3.5. Cytotoxic T-lymphocyte assays CTL responses were examined with PBMC drawn 15 weeks p.i. from three cats inoculated with 1250 TCID50, three cats inoculated with 50 TCID50 and two uninfected cats. After 14 days of in vitro stimulation of the PBMC with autologous FIV-PPR infected antigen presenting cells, detectable memory CTL responses were determined. The CTL responses were variable in both groups (Fig. 3a and b). All three cats in the 1250 TCID50 infected group had demonstrable CTL responses. The best CTL responses were observed

322

R.M. Hokanson et al. / Veterinary Microbiology 76 (2000) 311±327

Fig. 3. FIV-speci®c CTL responses 15 weeks p.i. with the PPR strain of FIV. The percent lysis refers to the amount of radioactive indium released from T lymphocyte target cells. The positive control effector and target cells were obtained from the PBMC of cat A306, that had been infected for 31 weeks with the PPR strain: (a) the CTL responses of cats inoculated with 1250 TCID50 of virus; (b) the CTL responses of cats inoculated with 50 TCID50 of virus.

from cat E238. With this cat, 50% of the target cells were lysed by T lymphocytes, with only a 50:1 effector to target ratio. Whereas cat E284 resulted in 10% or greater lysis depending on the target cells, the effector cells from cat E274 in this group demonstrated responses that were similar to uninfected cat controls (3% lysis at 100:1 E:T ratio). The effector cells from the cats given the lowest dose of virus were also variable in their capacity to lyse infected cells (Fig. 3b). Cat E290 responded at a 100:1 effector to target cell ratio with 20% lysis of target cells. The cat E254 effector cell responses were lower with 8% lysis at an E:T cell ratio of 50:1, one-fourth as great a response with half as many effector cells. The effector cells from infected cat E210 did not have detectable

R.M. Hokanson et al. / Veterinary Microbiology 76 (2000) 311±327

323

CTL activity at this time. PBMC from uninfected controls E266 and E244 did not have demonstrable CTL activity. Cat A306 infected in the earlier experiment had been shown to have reproducible CTL activity and therefore was used as a positive control. Cats A306 and A308 infected 4 months prior to the other animals in this study reproducibly had CTL activity of approximately 30 and 15% with an 80:1 E:T cell ratio. The CTL responses of E238 and E284 have reproducibly been approximately 50 and 7%, respectively (Choi et al., 2000). In these assays, the uninfected controls were 1.5% or less. 4. Discussion This study differs from others evaluating FIV infection in that virus was produced from a molecular clone, was cell-free and various doses of infectious virus were used. Outward general signs of illness in cats, such as anorexia, rough hair coat, and withdrawal/ depression were not observed in this study. Behavioral changes were also more severe in the animals receiving more virus than in those receiving the 50 TCID50 dose. Behavior changes persist with cat E238 (given the 1250 TCID50 dose) even two and half years p.i., whereas recovery from acute FIV infection and a protracted asymptomatic phase were expected in experimentally infected cats (Pedersen et al., 1987; Ackley et al., 1990, personal observations). Due to the SPF housing conditions, clinical signs of acute infection may be limited because exposure to secondary infectious agents that can exacerbate or accelerate illness was minimal (Dua et al., 1994). Overall, we demonstrated that illness was dose-dependent, with more clinical signs observed in the 250 and 1250 TCID50 groups and milder illness in the 50 TCID50 group. Elevated body temperatures in all the individuals within the 50 TCID50 group on day 12 may not be directly attributed to viral infection, but could be related to excited behavior. This group was particularly playful and energetic on day 12 p.i. From days 5±7, one animal in this group, E210, did have an elevated (>102.58C) body temperature which might be attributed to illness. Transient lymphadenopathy was dose-dependent, readily apparent in the 250 TCID50 and 1250 TCID50 infected animals, but not in cats receiving 50 TCID50 of PPR virus. Four additional animals have since been inoculated with the lowest dose without demonstration of lymphadenopathy or observable neurologic problems (unpublished data). The absence of lymphadenopathy described by Sparger et al. (1994) who quantitated virus by RT activity may have used relatively lower amounts of virus than the 250 and 1250 TCID50 doses used in the present study. In HIV infection, L-selectin expression on infected lymphocytes is unregulated (Wang et al., 1997). Expression of Lselectin increases homing of these cells to lymph nodes (Wang et al., 1997). Swelling of lymph nodes, due to the increased numbers of newly arrived lymphocytes, and subsequent localized reaction, due to the cellular branch of the immune system attempting to clear the infection, could help explain the transient nature of the early lymphadenopathy. The time when all individuals in the infected groups acquired detectable antibodies speci®c for the FIV capsid antigen was also dose-dependent. The greater lag time

324

R.M. Hokanson et al. / Veterinary Microbiology 76 (2000) 311±327

observed in the 50 TCID50 group is likely due to the lower initial levels of FIV in circulation p.i., thus, more time was needed to induce humoral immunity. Seroconversion within the ®rst 2 months p.i. has been noted for Petaluma (Pedersen et al., 1987; Yamamoto et al., 1988; Pancino et al., 1995), PPR (Pedersen et al., 1987; personal observations), NCSU1 (Tompkins et al., 1991; Callanan et al., 1992), and FIV-GLA-8 (Yu et al., 1998). A transient decrease in the CD4‡:CD8‡ T lymphocyte ratios was observed in all groups inoculated. The observed ratios were not dose responsive. Although Sparger et al. (1994) did not observe a modulation of CD4‡:CD8‡ T lymphocyte ratios in cats infected with the FIV-PPR molecular clone, it has been consistently observed following infection with virus (Pedersen and Barlough, 1991; English et al., 1994). Decreases in CD4‡ cell numbers have also been reported (Barlough et al., 1991; Pedersen and Barlough, 1991; Hoffmann-Fezer et al., 1992). Although not examined in this study, decreased responses to mitogens like Con A and pokeweed mitogen have also been reported (Barlough et al., 1991; Pedersen and Barlough, 1991). No signi®cant modulations were observed for the total number of lymphocytes. The increase in CD8‡ cells responding to infection may have been accompanied by a decrease in CD4‡ T lymphocytes. Therefore, unlike the humoral response, the response of CD8‡ T lymphocytes may not vary with the dose. The CTL responses were examined 15 weeks p.i. The results compared with the 6.4± 30% lysis observed after 21±30 days of in vitro stimulation using T cells from FIVPetaluma infected cats (Song et al., 1992). Interestingly, the effector cells from cat E284 lysed the mismatched target cells better than the autologous target cells. These studies more likely identi®ed memory CTL. Dose responsive effects on CTL activity could not be evaluated in effector cells stimulated with antigen presenting cells prior to determining cytolytic activity. The kinetics of CTL responses were not examined. Behavioral and motor dysfunctions support encephalopathic and central nervous system (CNS) disease, respectively, in cats examined. Behavioral alterations are the most common neurologic manifestation of feline lentiviral infection in cats (Dow et al., 1990; Podell et al., 1993). One of the 15 infected animals (in the 1250 TCID50 group) exhibited aberrant behavior that has progressively become more severe in the two and a half years p.i. Motor dysfunction is less commonly documented (Dow et al., 1990; Philips et al., 1996). Delayed pupillary and righting re¯exes, hindlimb paresis (weakness) and sleep disturbances are all reported signs of FIV induced neurologic abnormalities in experimentally infected cats (Philips et al., 1994, 1996; Prospero-Garcia et al., 1994; Vitkovic et al., 1995). Motor dysfunction may be a residual effect of FIV infections within the CNS as also suggested by the neurologic examination results of two chronically infected cats, A306 and A308, in the pilot study. CNS impairment was subjectively documented by the neurological examination and neuroanatomic localization. Other neurodiagnostic procedures involving electrophysiologic testing (electroencephalography, visual and auditory evoked potentials) and magnetic resonance imaging have been used to identify CNS abnormalities following FIV infections (Podell et al., 1993; Philips et al., 1996). Biweekly complete blood count evaluations did not reveal any changes of lymphocytes and neutrophils concentrations following infection with any dose of the PPR-strain. This

R.M. Hokanson et al. / Veterinary Microbiology 76 (2000) 311±327

325

is in contrast to experimental observations by Dua et al. (1994) and Zenger et al. (1993) using FIV-Petaluma, in which the lymphocyte and neutrophil numbers dropped within the ®rst 2±6 weeks p.i. Following HIV infection, the levels of these leukocytes are reported to fall during the AIDS-related-complex (ARC) and AIDS phases of infection in humans as described by Zaaijer et al. (1995). No differences were detected in the isolation of virus or the PCR ampli®cation of provirus from the cats receiving the various doses of PPR-strain virus. These procedures are not quantitative and therefore do not evaluate differences in the viral load. Viral isolation was more sensitive because considerably more cells were used in this assay than in the PCR. Although there were overlaps among the groups, the indicators of pathogenesis increased with increasing viral inocula, especially between the 50 and either the 250 or 1250 TCID50 doses. The 1250 TCID50 dose would be the most useful dose for evaluating consistent diagnostic and clinical illness that is indicative of infection. Observed signs of infection were milder in the 50 TCID50 group. The lowest dose may be more representative of most naturally occurring infections. Acknowledgements This work was supported by funds from Schering Plough Animal Health. Partial support was provided by grant R21AI43882-01 from the NIH National Institute of Allergy and Infectious Diseases and by grant No. 96FE-09 from the Morris Animal Foundation. The authors thank J. Jason Page, Drs. Michelle Aucoin, Jim Rugila and Li Wang, for their valuable technical assistance and animal care, and Dr. Zhongxia Li for his advise and comments. References Ackley, C.D., Hoover, E.A., Cooper, M.D., 1990. Identi®cation of a CD4 homologue in the cat. Tiss. Antigens 35 (2), 92±98. Barlough, J.E., Ackley, C.D., George, J.W., Levy, N., Acevedo, R., Moore, P.F., Rideout, B.A., Cooper, M.D., Pedersen, N.C., 1991. Acquired immune dysfunction in cats with experimentally induced feline immunode®ciency virus infection: comparison of short-term and long-term infections. J. Aquired Immune De®c. Syndr. 4 (3), 4219±4227. Bendinelli, M., Pistello, M., Lombardi, S., Poli, A., Garzelli, C., Matteucci, D., Ceccherini-Nelli, L., Malvaldi, G., Tozzini, F., 1995. Feline immunode®ciency virus: an interesting model for AIDS studies and an important cat pathogen. Clin. Micro Rev. 8 (1), 87±112. Brown, W.C., Bissey, L., Logan, K., Pedersen, N.C., Elder, J.H., Collisson, E.W., 1991. Feline immunode®ciency virus infects both CD4‡ and CD8‡ T-lymphocytes. J. Virol. 65 (6), 3359±3364. Burleson, F.G., Chambers, T.M., Wiedbrauk, D.L., 1992. Virology: A Laboratory Manual. Academic Press/ Harcourt, Brace and Jovanovich, San Diego, CA, pp. 244±245. Callanan, J.J., Thompson, H., Toth, S.R., O'Neil, B., Lawrence, C.E., Willett, B., Jarrett, O., 1992. Clinical and pathological ®ndings in feline immunode®ciency virus experimental infection. Vet. Immunol. Immunopath. 35, 3±13. Carpenter, M.A., Brown, E.W., Cullver, M., Johnson, W.E., Pecon-Slattery, J., Brousset, D., O'Brien, S.J., 1996. Genetic and polygenetic divergence of feline immunode®ciency virus in the puma (Puma concolor). J. Virol. 70 (10), 6682±6693.

326

R.M. Hokanson et al. / Veterinary Microbiology 76 (2000) 311±327

Choi, I.S., Hokanson, R., Collisson, E.W., 2000. Anti-feline immunode®ciency virus (FIV) soluble factor(s) produced from antigen-stimulated feline CD8‡ T lymphocytes suppresses FIV replication. J. Virol. 74, 676±683. Dow, S.W., Poss, M.L., Hoover, E.A., 1990. Feline immunode®ciency virus: a neurotropic lentivirus. J. Aquired Immune De®c. Syndr. 3, 658±668. Dua, N., Reubel, G., Moore, P.F., Higgins, J., Pedersen, N.C., 1994. An experimental study of primary feline immunode®ciency virus infection in cats and a historical comparison to acute simian and human immunode®ciency virus diseases. Vet. Immunol. Immunopath. 43 (4), 337±355. English, R.V., Nelson, P., Johnson, C.M., Nasisse, M., Tompkins, W.A., Tompkins, M.B., 1994. Development of clinical disease in cats experimentally infected with feline immunode®ciency virus. J. Infect. Dis. 170, 543±552. Hoffmann-Fezer, G., Thum, J., Ackley, C., Herbold, M., Mysliwietz, J., Thefeld, S., Hartmann, K., Kraft, W., 1992. Decline in CD4‡ cell numbers in cats with naturally acquired feline immunode®ciency virus infection. J. Virol. 66 (3), 1484±1488. Jarrett, O., 1996. The relevance of feline retroviruses to the development of vaccines against AIDS. AIDS Res. Human Retrov. 12 (5), 385±387. Li, J., Brown, W.C., Song, W., Carpino, M.R., Wolf, A.M., Grant, C.K., Elder, J.H., Collisson, E.W., 1995. Retroviral vector-transduced cells expressing the core polyprotein induce feline immunode®ciency virusspeci®c cytotoxic T-lymphocytes from infected cats. Virus Res. 38, 93±109. Matsumara, S., Ishida, T., Washizu, T., Tomada, I., Nagata, S., Chiba, J., Kurata, T., 1993. Pathologic features of acquired immunode®ciency-like syndrome in cats experimentally infected with feline immunode®ciency virus. J. Vet. Med. Sci. 55 (3), 387±394. Oliver, J.E., Lorenz, M.D., Kornegay, J.N., 1997. Neurologic history and examination. In: Oliver, J.E., Lorenz, M.D., Kornegay, J.N. (Eds.), Handbook of Neurobiology, 3rd Edition. Saunders, Philadelphia, PA, pp. 3±46. Osofsky, S.A., Hirsch, K.J., Zuckerman, E.E., Hardy Jr., W.D., 1996. Feline lentivirus and feline oncovirus status of free-ranging lions (Panthera leo), leopards (Panthera pardus), and cheetahs (Acinonyx jubatus) in Botswana: a regional prospectus. J. Zool. Wildlife Med. 27 (4), 453±467. Pancino, G., Castelot, S., Sonigo, P., 1995. Differences in feline immunode®ciency virus host cell range correlates with envelope fusogenic properties. Virology 206 (2), 796±806. Pedersen, N.C., Barlough, J.E., 1991. Clinical overview of feline immunode®ciency virus. J. Am. Vet. Med. Assoc. 199 (10), 1298±1305. Pedersen, N.C., Ho, E.W., Brown, M.L., Yamamoto, J.K., 1987. Isolation of a T-lymphotropic virus from domestic cats with an immunode®ciency-like syndrome. Science 235, 790±793. Philips, T.R., Talbott, R.L., Lamont, C., Muir, S., Lovelace, K., Elder, J.H., 1990. Comparison of two host cell range variants of infection of cats by injection with DNA of an feline immunode®ciency virus clone. Virology 238 (1), 157±160. Philips, T.R., Prospero-Garcia, O., Puaoi, D.L., Lerner, D.L., Fox, H.S., Olmsted, R.A., Bloom, F.E., Henrikson, S.J., Elder, J.H., 1994. Neurological abnormalities associated with feline immunode®ciency virus infection. J. Gen. Virol. 75, 979±987. Philips, T.R., Prospero-Garcia, O., Wheeler, D.W., Wagaman, P.C., Lerner, D.L., Fox, H.S., Whalen, L.R., Bhoom, F.E., Elder, J.H., Henrikson, S.J., 1996. Neurological dysfunction caused by a molecular clone of feline de®ciency virus, FIV-PPR. J. Neuro Virol. 2, 388±396. Podell, M., Oglesbee, M., Mathes, L., Krakowka, S., Olmstead, R., Lafrado, L., 1993. AIDS-associated encephalopathy with experimental feline immunode®ciency virus infection. J. Aquired Immune De®c. Syndr. 6 (7), 758±771. Poss, M.L., Dow, S.W., Hoover, E.A., 1992. Cell-speci®c envelope glycosylation distinguishes FIV glycoproteins produced in cytopathically and noncytopathically infected cells. Virology 188, 25±32. Prospero-Garcia, O., Herold, N., Phillips, T.R., Elder, J.H., Bloom, F.E., Henrikson, S.J., 1994. Sleep patterns are disturbed in cats infected with feline immunode®ciency virus. Proc. Natl. Acad. Sci. USA 91, 12947±12951. Song, W., Collisson, E.W., Billingsley, P.M., Brown, W.C., 1992. Induction of feline immunode®ciency virusspeci®c cytolytic T-cell responses from experimentally infected cats. J. Virol. 66 (9), 5409±5417. Sparger, E.E., Beebe, A.M., Dua, N., Himathogkam, S., Elder, J.H., Torten, M., Higgins, J., 1994. Infection of cats with molecularly cloned and biological isolates of the feline immunode®ciency virus. Virology 205 (2), 546±553.

R.M. Hokanson et al. / Veterinary Microbiology 76 (2000) 311±327

327

Sparger, E.E., Louie, H., Ziomeck, A.M., Luciw, P.A., 1997. Infection of cats by injection with DNA of a feline immunode®ciency virus clone. Virology 238 (1), 157±160. Tompkins, M.B., Nelson, P.D., English, R.V., Novotney, C., 1991. Early events in the immunopathogenesis of feline retrovirus infections. J. Am. Vet. Med. Assoc. 199 (10), 1311±1315. Vitkovic, L., Stover, E., Koslow, S.H., 1995. Animal models recapitulate aspects of HIV/CNS disease. AIDS Res. Human Retro. 11 (6), 753±759. Wang, L., Robb, C.W., Cloyd, M.W., 1997. HIV induces homing of resting T lymphocytes to lymph nodes. Virology 228 (2), 141±152. Yamamoto, J.K., Sparger, E., Ho, E.W., Andersen, P.R., O'Connor, T.P., Mandell, C.P., Lowenstine, L., Munn, R., Pedersen, N.C., 1988. Pathogenesis of experimentally induced feline immunode®ciency virus infection in cats. Am. J. Vet. Res. 49 (8), 1246±1258. Yu, N., Billaud, J.N., Phillips, T.R., 1998. Effects of feline immunode®ciency virus on astrocyte glutamate uptake: implications for lentivirus-induced central nervous system diseases. Proc. Natl. Acad. Sci. USA 95 (5), 2624±2629. Zaaijer, H.L., Kok, W., ten Veen, J.H., Reesink, H.W., Foolen, H., Winkel, I.N., Huisman, J.G., Cuypers, H.T., Kievits, T., Lelie, P.N., 1995. Detection of HIV-1 RNA in plasma by isothermal ampli®cation (NASBA) irrespective of the stage of HIV-1 infection. J. Virol. Meth. 52 (1/2), 175±181. Zenger, E., Brown, W.C., Song, W., Wolf, A.M., Pedersen, N.C., Longnecker, M., Li, J., Collisson, E., 1993. Evaluation of cofactor effect of feline syncytium-forming virus on feline immunode®ciency virus infection. Am. J. Vet. Res. 54 (5), 713±718.