Mouse Moloney Leukemia Virus Infects Microglia but Not Neurons Even
Total Page:16
File Type:pdf, Size:1020Kb
Molecular Psychiatry (1997) 2, 104–106 1997 Stockton Press All rights reserved 1359–4184/97 $12.00 CYTOKINES IN THE BRAIN fragment that encompasses the 3′ end of pol and all of env6 and more specifically to point mutations in the env gene that results in a four amino acid substitution Mouse Moloney leukemia in the ts1 MoMuLV envelope protein.7,8 Replacement of the ts1 MoMuLV paralytogenic gene virus infects microglia but fragment with a homologous fragment from parental not neurons even though it (wt) MoMuLV not only corrects the env processing defect, but prevents paralysis in inoculated mice.6 In induces motor neuron addition, the central nervous system (CNS) of ts1 MoMuLV-infected paralyzed mice in the terminal disease stages of disease contains significantly more Pr80env than that of age-matched mice infected with the nonpa- 1 2 1 JF Zachary , TV Baszler , RA French and ralytogenic (wt) MoMuLV.5,9 The concentration of KW Kelley3 Pr80env in the brain is directly proportional with higher concentrations of other viral proteins (p30, gp70), pre- 1 College of Veterinary Medicine, University of Illinois; sumably due to higher overall virus replication in the 2 College of Veterinary Medicine, Washington State brains of ts1 MoMuLV-infected mice.9 Together, these 3 University, Pullman, WA; College of Agriculture, University findings suggest that unprocessed Pr80env polyprotein of Illinois, Urbana, IL, USA may play a central role in the induction of neuronal dysfunction in the ts1 MoMuLV system. During disease progression, it has been documented that there is no Keywords: mice; Moloney murine leukemia virus; neuro- selective accumulation of unprocessed Pr80env within degenerative disease; nervous system; spongiform 9 encephalopathy; retrovirus infected microglial cells. The pathogenesis of neuro- logic disease in ts1 MoMuLV-infected mice is at Motor neuron degeneration caused by ts1 MoMuLV present undetermined but has been linked to: (1) the occurs by an indirect mechanism and hypothetically enhanced ability of the virus to replicate in the CNS;5 appears associated with a two-cell or three-cell patho- (2) a gene mutation and the defective processing and genesis hypothesis. The first step in this hypothesis is accumulation of the env precursor polyprotein, associated with a small subset of resident microglial env 7,8 cells that serve as the principal target cells for ts1 Pr80 ; (3) the thymus and selective T lymphocyte 10,11 MoMuLV infection. The second step is likely linked to populations; and (4) the selective infection of trophic events, probably mediated by cytokines, that microglial cells9,12 and possibly astrocytes.13 The lead to hypertrophy and activation of a substantial num- pathogenesis of ts1 MoMuLV-induced neuronal ber of additional microglial cells (autocrine effect) and degeneration has been partially elucidated by Baszler adjacent astrocytes (paracrine effect). The third step in and Zachary9,12 following intraperitoneal inoculation this hypothesis appears related to indirect neuronal of neonatal mice. degeneration mediated by cytotoxins produced by acti- Clinically, mice inoculated with ts1 MoMuLV vated microglial cells and astrocytes. In this last step, develop, within 24–48 h after birth, progressive paresis motor neurons located within these foci of activated and involuntary muscle tremors by day 15 post-inocu- microglial cells and astrocytes are ‘innocent bystander cells’ and degenerate and die due to paracrine effects. lation (p.i.) followed rapidly by weight loss, cachexia, The mechanism of motor neuron degeneration is poorly skeletal muscle atrophy, paralysis, and death by day 9,12,14 understood but is likely linked to a sequential cascade 35 p.i. Following intraperitoneal inoculation of of trophic factors and cytokines resulting in a final com- newborn mice, the virus initially and predominantly mon pathway for motor neuron death involving pro- replicates in megakaryocytes of the spleen and to a duction of oxidative radicals, excitatory aminoacid neur- lesser extent in lymphoid tissue of the spleen and thy- otransmitter-like substances, prostaglandins, or nitric mus.9,12 Virus is transferred hematogenously to the oxide. central nervous system within platelets or as a cell-free Moloney murine leukemia virus (MoMuLV), an viremia. In sequence, ts1 MoMuLV infects and repli- exogenous mammalian type C murine retrovirus, is an cates in capillary endothelial cells; spreads to, infects, E-tropic, Fv1-NB tropic, and non-paralytogenic retro- and replicates in perivascular microglial cells and per- virus1 that induces thymic or splenic T cell lymphomas icytes; and finally infects resident microglial cells.9,12 in mice and rats.2 Ts1 MoMuLV, a spontaneous tem- During this 35-day period, mice infected with ts1 perature-sensitive mutant of wild type (wt) MoMuLV,3 MoMuLV have a progressive loss of body (cachexia), causes a noninflammatory spongiform motor neuron thymic (atrophy), and splenic (atrophy) weights when disease in susceptible strains of newborn mice.4 The compared to control mice (JF Zachary, unpublished virus is defective in processing env precursor polypro- data).14 There are no differences in plasma corticos- tein Pr80env into gp70 and p15E, which results in intra- terone concentrations between ts1 MoMuLV-infected cellular accumulation of Pr80env in infected cells, and and control mice when measured by radioimmunoas- leads to the production of virions with reduced say at 5-day intervals from day 1 p.i. to day 35 p.i. (JF amounts of gp70 and p15E when compared to wt Zachary, unpublished data). TNF-a is not detected in MoMuLV virions.5 The gene segment responsible for plasma of ts1 MoMuLV-infected or control mice when the processing defect has been mapped to a 1.6-kb pair measured by immunobased assays at 5-day intervals Mouse Moloney leukemia virus JF Zachary et al 105 from day 1 p.i. to day 35 p.i. (JF Zachary, unpublished istic of an activated status.17,19 Cultures infected with data). Results from immunobased assays for plasma wt MoMuLV or incubated with culture media show no concentrations of IL1b and IL6 are pending. morphologic changes. Approximately 10–20% of the There are no gross lesions in the nervous system.5,14 ameboid microglial cells are infected with virus when Microscopic lesions include progressive degeneration qualitatively assessed by immunohistochemical analy- and loss of upper motor neurons in extrapyramidal sis or electron microscopy.17 Biological (WEHI) and nuclei of the brainstem (bulbar), degeneration and loss immunobased assays for TNF-a demonstrate elevated of lower motor neurons in cervical and lumbar enlarge- concentrations of this cytokine in cell culture super- ments of the spinal cord (anterior/ventral gray horn), natants following infection with ts1 MoMuLV or incu- and hypertrophic microgliosis and astrogliosis.14 bation with LPS.17 Analyses of wt MoMuLV-infected Lesions, which progress in severity, appear in specific microglial cell culture supernatants document much nuclei and areas of the brain and spinal cord at specific smaller increases in TNF-a concentrations; no TNF-a times during the course of the disease. This finding is present in supernatants from microglial cell cultures suggests receptor-mediated specificity for viral infec- incubated with culture media alone.17 Similar studies tion of target areas within the nervous system. The are in progress to determine if ts1 MoMuLV-infected sequence of histologic and ultrastructural lesions con- microglial cells produce IL1b and IL6. A nitrite assay sists of early astrocytic swelling; cytoplasmic swelling for reduced nitric oxide failed to demonstrate nitrite in followed by hydropic (vacuolar) degeneration of neu- supernatants from ts1 MoMuLV-infected microglial rons, astrocytes, and oligodendrocytes; neuronal loss cell cultures or from control cultures.17 Primary mixed (probable apoptosis); hypertrophic microgliosis; and glial cell cultures produced measurable quantities of fibrillary hypertrophic astrogliosis.9,12,14 Wt MoMuLV nitrite following infection with ts1 MoMuLV. This infection, which is subclinical and nonlethal, produces finding suggests that there is an important interaction the same patterns of lesion character, temporal devel- between microglial cells and astrocytes in the synthesis opment, and spatial distribution and infects the same of iNOS following infection and activation of target cells but the lesions are much less severe and microglial cells.17 Co-culture studies examining cyto- resolve with time when compared to ts1 MoMuLV- kine-induced neurocytotoxicity using neuron cell lines infected mice.14 (N1E-C1300) and ts1 MoMuLV-infected microglial Resident microglial cells are the major reservoir and cells are planned in the near future.17 Long-term stud- amplifier of ts1 MoMuLV infection in the CNS and ies will examine the role of excitotoxins such as gluta- neuronal degeneration is mediated indirectly by a mate and NMDA receptor-mediated cytotoxicity in the selective ts1 MoMuLV-induced activation of these resi- pathogenesis of ts1 MoMuLV-induced motor neuron dent microglial cells.9,12,15 There is a coinciding disease. hypertrophy of astrocytes in the areas of microglial cell proliferation (hypertrophy) and neuronal degener- ation.9 Of the large number of activated and hypertro- References phied microglial cells in these lesions, data suggest that 1 Gardner MB. Type-C viruses of wild mice: characteriz- approximately 10–20% of these cells are infected with ation and natural history of amphotropic, ectropic, and virus