Pseudorabies virus (PRV) has been used extensively to map synaptic circuits in the CNS and PNS. A fundamental assumption of these studies is that the virus replicates within synaptically linked populations of neurons and does not spread through the extracellular space or by cell-to-cell fusion. In the present analysis we have used electron microscopy to characterize pathways of viral replication and egress that lead to transneuronal infection of neurons, and to document the non-neuronal response to neuronal infection. Three strains of PRV that differ in virulence were used to infect preganglionic motor neurons in the dorsal motor nucleus of the vagus (DMV). The data demonstrate that viral replication and transneuronal passage occur in a stepwise fashion that utilizes existing cellular processes, and that the non-neuronal response to infection serves to restrict nonspecific spread of virus by isolating severely infected neurons. Specifically, capsids containing viral DNA replicate in the cell nucleus, traverse the endoplasmic reticulum to gain access to the cytoplasm, and acquire a bilaminar membrane envelope from the trans cisternae of the Golgi. The outer leaf of this envelope fuses with the neuron membrane to release virus adjacent to axon terminals that synapse upon the infected cell. A second fusion event involving the viral envelope and the afferent terminal releases the naked capsid into the bouton. Systematic analysis of serial sections demonstrated that release of virus from infected neurons occurs preferentially at sites of afferent contact. Nonspecific diffusion of virus from even the most severely infected cells is restricted by astrocytes and other non-neuronal elements that are mobilized to the site of viral infectivity. The ability of glia and macrophages to restrict spread of virus from necrotic neurons is the product of (1) temporal differences in the mobilization of these cells to the site of infection, (2) differential susceptibility of these cells to PRV infection, and (3) abortive viral replication in cells that are permissive for infection. The findings provide further insight into the intracellular routes of viral assembly and egress and support the contention that transneuronal spread of virus in the brain results from specific passage of virions through synaptically linked neurons rather than through cell fusion or release of virus into the extracellular space.

We examined the responses of astrocytes, ramified microglia, and brain macrophages to CNS neuronal infection with virulent or attenuated strains of a swine alpha herpesvirus (pseudorabies virus, PRV). After PRV inoculation of the rat stomach or pancreas, the temporal course of viral replication and induced pathology of infected neurons were assessed in the dorsal motor nucleus of the vagus (DMV) and amygdala using an antiserum generated against PRV. Specific monoclonal antibodies against glial fibrillary acidic protein (GFAP), OX42, and ED1 and morphological criteria were used to classify non-neuronal cells. Both PRV strains infected DMV and motor neurons and then passed transneuronally to infect brainstem neurons that innervate the DMV. However, the onset of neuronal infection produced by the attenuated strain occurred approximately 20 hr later than infection with the virulent strain. Animals infected with the attenuated strain also survived longer, permitting transneuronal passage of virus into forebrain areas of the visceral neuraxis. Neuronal infection with both PRV strains produced consistent alterations in astrocytes, ramified microglia, and brain macrophages that correlated spatially and temporally with progressive stages of viral replication and neuronal pathology. Early stages of infection were characterized by increases in immunoreactivity for astrocytic GFAP and microglial OX42 that preceded overt signs of neuronal pathology. At later stages, GFAP immunoreactivity decreased dramatically in focal areas of neuronal infection while OX42 immunoreactivity continued to increase. Subsequently, ED1-immunoreactive brain macrophages infiltrated these infected areas. Double immunocytochemical labeling demonstrated that some astrocytes and brain macrophages were immunopositive for viral antigens but ramified microglia were not. The responses of glia and brain macrophages are consistent with a proposed role in restricting extracellular spread of virus by isolating or phagocytosing infected cells. These phenomena may contribute to the specific transneuronal transport exhibited by PRV.

The alpha herpesviruses, a subfamily of the herpesviruses, are neurotropic pathogens found associated with most mammalian species. The prototypic member of this subfamily is herpes simplex virus type 1, the causative agent of recurrent cold sores in humans. The mild nature of this disease is a testament to the complex and highly regulated life cycle of the alpha herpesviruses. The cellular mechanisms used by these viruses to disseminate infection in the nervous system are beginning to be understood. Here, we overview the life cycle of alpha herpesviruses with an emphasis on assembly and transport of viral particles in neurons.

The neurotropic alpha-herpesviruses are common mammalian pathogens that invade the peripheral and central nervous system of their hosts. Their ability to invade and spread in the nervous system in a directional manner has been exploited to develop them as neuronal circuit tracers. Tracing viruses spread among synaptically connected neurons and, by assaying brain sections for viral antigen or reporter genes expressed from the viruses, chains of synaptically connected neurons can be visualized. Virulent field strains generally are not good tracers, but some attenuated strains perform well. Live attenuated vaccine strains of pseudorabies virus (PRV), such as PRV Bartha, are among the most popular virus circuit tracers. It may be counterintuitive that attenuation results in improved neural tracing that requires extensive replication and spread in the brain. This report summarizes two lines of experiments directed to resolving this apparent paradox and introduces a new paradigm for tracing viruses.

No abstract