Cns Targets For Hiv Infection

Mononuclear Phagocytes

The pathogenesis of HIV-associated CNS disease centers around the macrophage. Macrophages are the principal cell infected in the brain and become activated and recruited into tissue during inflammation and emigrate into the CNS during disease (Koenig et al., 1986; Schrier et al., 1993; Gabuzda and Wang, 1999; Gartner and Liu, 2002). This influx is transient, however, and will revert to a quiescent state after the inflammatory process has subsided. For HIVE and HAD, the process never subsides, as brain inflammation is continuous and induced by ongoing viral replication. One population of MP, meningeal macrophages, is characteristically infected early after the initial seroconversion reaction, paralleling the development of aseptic meningitis. Later in the course of the disease perivascular macrophages and microglia are infected preferentially (Kure et al., 1990; Devadas et al., 2004).

The perivascular macrophage is an actively studied MP cell type involved in HAD pathogenesis. In natural conditions, these cells exist between the glia limitans and basement membrane of the choroid plexus and CNS capillaries. Perivascular macrophages are derived from circulating monocytes but will not become fully active macrophages. They are in close association with the bone microvascular endothelial cells

(BMVEC), and this position allows them to serve as sentinels for the CNS. They are, in fact, intermediates between the circulation and the microglia. Since mi-croglia are in contact with these macrophages, signals may be rapidly communicated deep into the CNS from interactions at the perivascular space. Transmission of virus and/or inflammatory responses in the brain may occur between these perivascular macrophages and glial cells (Pulliam et al., 1991; Gendel-man et al., 1994; Kaul et al., 2001; Williams and Hickey, 2002; Persidsky and Gendelman, 2003; Devadas et al., 2004; Gonzalez-Scarano and MartinGarcia, 2005).

Current data strongly suggest that perivascular macrophages are likely responsible for most of the transmission of virus into the CNS (Pulliam et al., 1991; Gartner and Liu, 2002; Williams and Hickey, 2002; Devadas et al., 2004). Several observations support this hypothesis, including findings of perivascular macrophages infected with virus, often at high levels (Rappaport et al., 1999; Williams and Hickey, 2002). Through such cells, virus can be readily transferred to microglia upon microglial activation, since there is close contact between the two. The perivascular macrophages are the critical CNS-resident MP acting as an antigen-presenting cell to T cells (Tyor et al., 1992). Thus, they are at high risk for exposure and contact with infected T cells and/or inducing T-cell protective immunity. Moreover, there is relatively frequent turnover of perivascular macrophages, compared with that of microglia (Kennedy and Abkowitz, 1997; Ghorpade et al., 1998). They may bring virus into the CNS after being infected in the periphery, and release or transmit virus after cell death or through interactions with T cells or microglia.

Parenchymal microglia occur in significant numbers in the CNS and may constitute up to 10% of CNS cells. They enter the CNS during gestation and have a very low turnover rate (Kaur and Ling, 1991; Alliot et al., 1999). There are two morphological subtypes of microglia (Ling, 1982a). Ramified mi-croglia are resting cells with reduced secretory and phagocytic activity (Ling, 1982a; Glenn et al., 1992). They make up the web of microglia that spans the CNS. In contrast to perivascular macrophages, they have weak antigen-presenting capability. The second morphological subtype, amoeboid in form, is a morphological intermediate and transitional cell between the ramified microglia and the brain macrophages. This subtype is not found in the normal adult CNS

but rather in inflammatory and demyelinating conditions (Ling, 1982b; Kaur et al., 1985; Ling and Wong, 1993).

Infection of brain MPs leads to formation of multinucleated giant cells (MGC). These cells result from the fusion ofHIV-1-infected brain MP with uninfected monocyte-derived macrophages (MDM) or microglia (Budka, 1986, 1991; Pontow et al., 2004). This fusion is mediated by HIV-envelope glycoproteins present at the surface of infected cells with CD4 and chemokine receptors at the surface of uninfected cells (Dalgleish etal., 1984; Lifsonetal., 1986a, 1986b; Matthews etal., 1987). The MGC are large, irregularly round, elongated or polyhedral, with dense eosinophilic cytoplasm in the center and vacuolated at the periphery (Budka et al., 1988; Pontow et al., 2004). Giant cell formation is found throughout the brain in HIV disease, but is characteristically seen primarily in the deep brain structures and most commonly in subcortical white matter. Although pathognomonic of HAD, giant cells are only found in 50% of patients (Sharer et al., 1985, 1988; Dickson, 1986).

Moreover, CNS macrophages are protected from most antiviral medication in part because of the blood-brain barrier (BBB). The nature of this viral reservoir and its abilities to harbor and support virus growth remain a key obstacle to eradication of HIV-1 infection in the CNS.


Astrocyte function, critical for the survival of neurons, may be impaired in the context of HIV-1 infection. Astrocytes are responsible for maintaining homeosta-sis in the CNS and are important in the detoxification of excess excitatory amino acids such as extracellular glutamate levels (Wesselingh and Thompson, 2001; Deshpande et al., 2005). However, infected astrocytes can produce cellular factors that may adversely affect neuronal survival (Lawrence and Major, 2002). Astrocytes play a dual role in the pathogenesis of HIV-related encephalopathy. In HIV-1 infection, astrocyte glutamate reuptake is impaired, possibly due to interactions with infected macrophages (Fine et al., 1996; Jiang et al., 2001). In addition, glutamate release from the astrocyte is induced by activated macrophages (Vesce et al., 1997; Bezzi et al., 2001). Activation of the CXCR4 receptor by stromal cell-derived factor 1 (SDF-1) results in the release of extracellular TNF-a and downstream release of glu tamate (Bezzi et al., 2001). During HIV-1 infection there is an amplification or regulation of neurotoxic signals among astrocytes and microglia (Genis et al., 1992; Nottet et al., 1995). The HIV protein Tat induces expression in astrocytes of MCP-1, a chemoat-tractant for macrophages, and IL-8 and inducible protein-10 (IP-10), which attract multiple leukocyte types (Conant et al., 1998; Kutsch et al., 2000).

Astrocytes both proliferate and undergo apoptosis in HIV-1 CNS infection (Wesselingh and Thompson, 2001). The level of astrocyte apoptosis correlates strongly with both the severity and rate of progression of HIV dementia. Astrocytes can be infected (Wiley, 1986) in the absence of classical CD4 receptor through other chemokine receptors (Lawrence and Major, 2002), such as CXCR4 (Bezzi et al., 2001). Astrocytes may also serve as a reservoir for virus. In contrast to primary infection in MP, HIV-1 infection of astrocytes is not considered productive. The molecular events that limit HIV infection in primary astrocytes have been attributed to the inefficient translation of HIV structural proteins (Wesselingh and Thompson, 2001). The actual percentage of restrictively infected astrocytes in brains of patients with HIVE is unknown but thought to be relatively small.

T Cells

Impaired immune response is characteristic of late stages of HIV-1 neurodegeneration and HAD patho-genesis. This could be as a result of infection of CD4+ T cells, which occurs frequently and in the early stage of disease. T-cell abnormalities such as CD4+ T-cell lymphopenia, characterized by decreased lymphoproliferation and a decreased number of cells having a naïve phenotype, are seen in HIV-1-infected individuals (Devadas et al., 2004). An uninfected individual normally has 800 to 1200 CD4+ T cells per cubic millimeter of blood, but during HIV infection these cell numbers progressively decline. In addition, infiltrating CD8+ cells lose their protective role in later stages of infection, ultimately exhibiting impaired cytokine production and cytolysis, possibly as a result of anergy and the inability to eliminate HIV-1-infected cells in the setting of functionally impaired helper CD4+ T lymphocytes (Lewis et al., 1994; Liegler and Stites, 1994; Oxenius et al., 2004).

CD4+ T cells also serve as important reservoirs of HIV; a small proportion of these cells harbor HIV in a stable, inactive form. Normal immune processes may activate these cells, resulting in the production of new HIV virions. Once past the BBB, T cells are able to instigate cell-to-cell spread of HIV through CD4-mediated fusion of an infected cell with an uninfected cell. In addition, phagocytosis of CD4+ T cells by MP can result in the spread of virus (Budka, I99I; Lima et al., 2002). Activated T cells penetrate the BBB after insult to the CNS and can initiate both protective and toxic inflammatory responses (Petito et al., I986). Protective responses are elicited through elimination of the ongoing infectious agent by innate, humoral, and cytotoxic immune activities. Nonetheless, widespread inflammation in the setting of HIV often leads to damage of the BBB and further transendothelial migration of leukocytes entering the nervous system (Persidsky et al., 2000). Inflammation of the brain and spinal cord actively attracts T cells to the CNS. Macrophage inflammatory protein-I alpha (MIP-Ia) and MIP-I ß are relevant to the cellular recruitment and immune activation during HIV infection (Canque et al., I996; Jennes et al., 2004), as both use CCR5 as their receptor (Farzan et al., I997; Navenot et al., 200I; Miyakawa et al., 2002). MIP-Ia selectively attracts CD8+ and MIP-Iß recruits CD4+ lymphocytes. Both MIP-Ia and MIP-I ß are produced by HIV-I-infected monocytes and are closely linked to viral replication (Schmidtmayerova et al., I996).

During HIV-I infection, moreT cells are activated to a blast phase (Marcondes et al., 200I; Chen et al., 2005; Holm and Gabuzda, 2005). Once within the CNS, the lymphoblasts search for antigen as they migrate through the parenchyma. Such cells can easily encounter and engage perivascular macrophages through direct cell-to-cell contact or through soluble factors released. CD4+ T lymphocytes are responsible for most of the HIV replication in the periphery. HIV-I may enter the CNS in infected lymphocytes during the late stage of the disease. As the T cells migrate through the parenchyma, they secrete the cytokines that lead to activation ofMP and an amplification ofinflammatory cell responses throughout the CNS region involved (Weidenheim et al., I993). If these activated T cells are infected, they will shed virus as they migrate. At the same time, they induce CD4 expression on cells susceptible to HIV infection, rendering them even more susceptible. T cells expressing the CD40 ligand (soluble and bound forms) can activate both infected and noninfected monocytes that express TNF-a and CD40 receptors (Kornbluth et al., I998; Zhang et al., 2004; Chen et al., 2006). Macrophages become activated by way of scavenger receptors as they clear the debris of dead virus-infected cells (Lima et al., 2002, 2006). Since viral replication occurs mostly within CD4+ T cells, direct cytopathic effects of HIV-1 may be attributed to cell death (Zhang etal., 1999). The decline in CD4+ T lymphocytes allows macrophages, without control, to express a metabolically active, tissue-destructive phenotype.

0 0

Post a comment