CELLULAR MECHANISMS OF FELINE IMMUNODEFICIENCY VIRUS (FIV)-INDUCED NEUROPATHOGENESIS

Elisabeth Zenger, Evelyn Tiffany-Castiglioni, and Ellen W. Collisson

Department of Medicine, UCSF AIDS Program, San Francisco General Hospital, University of California, San Francisco, California (Zenger) and the Departments of Veterinary Anatomy and Public Health (Tiffany-Castiglioni) and Pathobiology (Collisson), College of Veterinary Medicine, Texas A&M University, College Station, Texas.

Received 10/7/97 Accepted 10/23/97

5. PERSPECTIVES

The lentiviruses HIV and FIV are able to efficiently enter the CNS and cause primary neurological disease. The characterization of the early stages of CNS infection by HIV, and its progression to the terminal phase, are fundamental in understanding physiopathology, and can only be analyzed in a non-human model. The FIV-infected cat serves as an excellent animal model for human AIDS and AIDS-associated neurological abnormalities. In vitro models of viral infection are of great interest as a means of studying possible mechanisms by which lentiviruses may impair neural cells, although these models cannot reproduce all cellular interactions within the CNS in vivo. This review describes several in vitro experiments designed to investigate the effects of FIV at the cellular level.

Although the exact role of astroglia in the neuropathogenesis of NeuroAIDS is yet to be clearly established, results of experiments described herein suggest that altered astroglial function may likely relate to the neurologic signs observed. For both FIV and HIV infections, there is a paradox between the magnitude of clinical signs observed and the paucity of CNS lesions. The converse situation is also seen. Several authors have commented on the surprising lack of lesions based on the severity of dementia or the lack of dementia based on the severity of lesions with human NeuroAIDS. Study of FIV strains that fit these extremes of clinical expression may contribute to the understanding of this paradox. Differences in neurovirulence observed in vivo may relate to specific differences in the mechanism of astroglial infection, viral expression, tendency to cause cytotoxicity, and ability to cause alteration of astroglial function.

Certain viruses which have the ability to persistently infect neural cells spare host cell vital functions but subtly interfere with its ability to perform specialized functions (69). Based on findings discussed, the pattern of astroglial infection by FIV varies with the strain being studied. Petaluma strain FIV, which is associated with significant neuropathology but minimal to no neurologic deficits, is cytotoxic to astroglia. Maryland strain FIV, which is associated with profound neurologic deficits, but minimal neuropathology, does not cause altered cell morphology or detectable cytotoxicity. It is possible that FIV-Pet infection in the CNS is arrested by rapid cell death of infected astroglia and ensuing inflammation, whereas FIV-MD may trigger a process or processes that, once initiated, may lead to disturbances in CNS homeostasis, eventually resulting in disease. One can expect that FIV-MD may thusly perturb complex interactions between astrocytes and neurons.

As information has accrued, it is becoming increasingly evident that the relationship between neurological signs and lentiviral infection of the CNS are not straight forward and cannot be explained by simple cytolytic brain infection. The observed neurologic dysfunction is likely multifactorial and complex, involving an intricate web of subcellular pathways and neurotoxic factors interacting with microglia, astroglia, and neurons. The importance of astroglia in the maintenance of CNS functions suggests that virally induced changes in infected astroglia may be essential in the progression of NeuroAIDS. A substantial degree of neuronal loss can occur in the cortex of HIV- or FIV-infected patients. There are three ways that neurons may die following lentivirus infection of the brain: murder, manslaughter, or suicide. Based on the accumulated evidence, murder is not very likely but one could make the case for manslaughter by substandard excitatory neurotransmitter uptake or suicide in the face of an unfavorable environment.

Among the many unresolved issues about lentivirus-mediated neuropathogenesis is how the lentivirus is able to penetrate an intact BBB. Despite definitive evidence, the most widely held belief is that the HIV and FIV gain access into the brain via infected hematogenous cells. Although the BBB is generally a formidable barrier to cells in circulation, activated lymphocytes readily traffic through the CNS. The cell-cell adhesion described in section 4.1. may be cytokine driven and integrin mediated, although neither of these factors have been adequately investigated because of limited reagent availability. The results of the experiments examining the cell-cell interaction and virus transmission, albeit mostly negative, provide fascinating suggestions to further explore the mechanism of brain infection by neurovirulent lentiviruses.

Astrocytes are crucial to the normal homeostatic regulation of the neuronal microenvironment, in large part because of their ability to selectively regulate extracellular levels of glutamate (70). Impaired astroglial uptake of glutamate following exposure to FIV suggests a mechanism of neurotoxicity whereby FIV infection may promote an increase in extracellular glutamate within the brain, thereby leading to neuronal excitotoxicity and increased neuronal sensitivity to oxidative stress. There was a difference noted in the severity of the glutamate uptake defect between the two strains of virus studied. Petaluma strain FIV is associated with much more profound drop in glutamate uptake whereas FIV-MD is associated with a subtle drop. The typical morphologic characteristics of excitotoxic injury (i.e., cell swelling) are consistent with a necrotic type of death. However, recent studies have provided evidence that exposure of neurons to relatively short durations or low concentrations of NMDA induces a delayed form of neurotoxicity predominated by apoptotic features (71). Brains from cats infected with FIV-Pet consistently demonstrate microfoci of inflammation and gliosis on postmortem examination (4), a finding that is consistent with the scenario of excessive extracellular glutamate and associated neuronal necrosis. Brains from cats infected with FIV-MD display very subtle postmortem lesions consisting primarily of white matter pallor (6, 9). This lesion is consistent with neuronal death without inflammation, a situation compatible with apoptosis. These proposed scenarios are certainly quite speculative and require much study for confirmation. However, results of these experiments are at least provocative and may provide a thread with which we may begin the unraveling of this obviously very complex mystery.