Puma lentivirus is controlled in domestic cats after mucosal exposure in the absence of conventional indicators of immunity

Abstract

A high percentage of free-ranging pumas (Felis concolor) are infected with feline lentiviruses (puma lentivirus, feline immunodeficiency virus Pco [FIV-Pco], referred to here as PLV) without evidence of disease. PLV establishes productive infection in domestic cats following parenteral exposure but, in contrast to domestic cat FIV, it does not cause T-cell dysregulation. Here we report that cats exposed to PLV oro-nasally became infected yet rapidly cleared peripheral blood mononuclear cell (PBMC) proviral load in the absence of a correlative specific immune response. Two groups of four specific-pathogen-free cats were exposed to PLV via the mucosal (oro-nasal) or parenteral (i.v.) route. All animals were PBMC culture positive and PCR positive within 3 weeks postinfection and seroconverted without exhibiting clinical disease; however, three or four oro-nasally infected animals cleared circulating proviral DNA within 3 months. Antibody titers reached higher levels in animals that remained persistently infected. PLV antigen-induced proliferation was slightly greater in mucosally inoculated animals, but no differences were noted in cytotoxic T-lymphocyte responses or cytokine profiles between groups. The distribution of virus was predominantly gastrointestinal as opposed to lymphoid in all animals in which virus was detected at necropsy. Possible mechanisms for viral clearance include differences in viral fitness required for crossing mucosal surfaces, a threshold dose requirement for persistence, or an undetected sterilizing host immune response. This is the first report of control of a productive feline or primate lentivirus infection in postnatally exposed, seropositive animals. Mechanisms underlying this observation will provide clues to containment of immunodeficiency disease and could prompt reexamination of vaccine-induced immunity against human immunodeficiency virus and other lentiviruses.

FIG. 1.

Timeline of study events. Blood samples were taken very frequently early postinoculation to document initial infection, seroconversion, and cytokine profiles. Once infection was established sampling frequency was decreased, since parameters being monitored attained steady state.

FIG. 2.

Lymphocyte subsets are not altered during PLV infection. (Top) Total CD4⁺ count. Lymphocyte numbers significantly declined only transiently after inoculation with PLV by the i.v. route. There was no decline in CD4⁺ lymphocytes in the mucosally inoculated group. (Bottom) Total CD8⁺ count. Lymphocyte numbers did not change in the i.v.-inoculated group, but in the mucosally inoculated group they were significantly increased on day 35 and decreased on day 107. Symbols represent the means of four animals. Bars represent 1 standard deviation. Asterisks indicate statistically significant changes (P < 0.05) from preexposure samples.

FIG. 3.

Proviral load is dependent upon route of infection and was cleared following mucosal infection in three of four animals. Initial proviral load was lower in three of four mucosally inoculated cats (B) than in i.v.-inoculated cats (A). Provirus subsequently dropped below the limit of detection in these three animals. No virus was detected in nonexposed control animals. The control line indicates the lower limit of detection for the assay. This lower threshold decreased after day 65 due to a decrease in DNA elution volume and consequent enhanced assay sensitivity.

FIG. 4.

PLV-specific antigen proliferation does not correlate with proviral load. Proliferation was only significantly elevated at day 49 (asterisk) post-virus exposure in the mucosally inoculated group. Individual responses are shown over time in the i.v.-inoculated (A) and mucosally inoculated (B) groups.

FIG. 5.

Cytokine profiles are highly variable following PLV infection and do not demonstrate a significant shift toward a TH-2 response. (A) IL-12/IL-10 ratio over time. (B) IFN-γ/IL-10 ratio. Symbols represent the mean values, and bars represent 1 standard deviation. Asterisks indicate significant (P < 0.05) differences from control group.

FIG. 6.

Steroid treatment does not result in increased PBMC proviral burden. Eight animals which had received PLV by either a mucosal (M1 to M4) or a parenteral (IV1 to IV4) route were treated with prednisone for 5 days. Blood samples were obtained 3 days prior to prednisone treatment and on days 5, 12, 19, and 26 after treatment was withdrawn. (A) Mean neutrophil counts indicated a typical steroid-induced neutrophilia occurred. Lines above the bar represent 1 standard deviation. (B) PBMC proviral load in individual animals. Virus was below the level of detection in three animals (M1, M2, and M3) before and after steroid treatment. Of the remaining five animals (IV1 to IV4 and M4), proviral load decreased slightly in three animals and increased slightly in two animals during steroid treatment. Signal was not detected in control animals.

FIG. 7.

PLV infection is enterotropic. Higher levels of provirus were found in the gastrointestinal tract and draining lymphatic tissue (B) than in lymphoid and other tissues (A) (note differences in scales). Bars represent the mean of all positive samples for each group reported. Lines above the bar represent 1 standard deviation. Numbers above the bars indicate the proportion of animals in the group in which virus was detected in the tissue sample. PBMC proviral burden at week 36 is also shown for comparison. No provirus was detected in control tissues obtained from PLV-negative cats.