Increased dopaminergic neurotransmission in therapy-naïve asymptomatic HIV patients is not associated with adaptive changes at the dopaminergic synapses.

Central dopaminergic (DA) systems are affected during human immunodeficiency virus (HIV) infection. So far, it is believed that they degenerate with progression of HIV disease because deterioration of DA systems is evident in advanced stages of infection. In this manuscript we found that (a) DA levels are increased and DA turnover is decreased in CSF of therapy-naïve HIV patients in asymptomatic infection, (b) DA increase does not modulate the availability of DA transporters and D2-receptors, (c) DA correlates inversely with CD4+ numbers in blood. These findings show activation of central DA systems without development of adaptive responses at DA synapses in asymptomatic HIV infection. It is probable that DA deterioration in advanced stages of HIV infection may derive from increased DA availability in early infection, resulting in DA neurotoxicity. Our findings provide a clue to the synergism between DA medication or drugs of abuse and HIV infection to exacerbate and accelerate HIV neuropsychiatric disease, a central issue in the neurobiology of HIV.

Dopamine deficits and regulation of the cAMP second messenger system in brains of simian immunodeficiency virus-infected rhesus monkeys.

The basal ganglia, structures rich in the neurotransmitter dopamine, are primarily affected during human immunodeficiency virus (HIV) infection. The authors measured levels of dopamine and its metabolites, homovanillic acid and 3,4-dihydroxyphenylacetic acid, in brains of uninfected and simian immunodeficiency virus (SIV)-infected rhesus monkeys during the asymptomatic stage of the infection. Moreover, the authors investigated changes in cyclic adenosine monophosphate (cAMP) and cAMP response element-binding protein (CREB), two factors involved in the signaling pathway of dopamine. The brain regions examined were the nucleus accumbens and the corpus amygdaloideum, which are limbic structures of the basal ganglia that are involved in the pathophysiology of psychiatric disorders and substance abuse. Dopamine content was reduced in both regions of SIV-infected monkeys compared to uninfected animals. Moreover, dopamine deficits were associated with a decrease in expression of total CREB. Intracellular concentrations of cAMP were decreased in nucleus accumbens and remained unchanged in corpus amygdaloideum of SIV-infected macaques. Changes in dopamine signaling were not related to pathology or viral load of the investigated animals. The results suggest that dopamine defects precede neurologic deficits and implicate dysfunction of the dopaminergic system in the etiopathogenesis of HIV dementia. Therefore, affective complications in HIV subjects should not be interpreted only as reactive psychological changes. The alterations in the mesolimbic dopaminergic system during asymptomatic stage of SIV infection implicate a biological background for psychiatric disorders in HIV infection.

Modulation of simian immunodeficiency virus neuropathology by dopaminergic drugs.

Drug abuse and human immunodeficiency virus (HIV) infection seem to cause cumulative damage in the central nervous system (CNS). Elevated extracellular dopamine is thought to be a prime mediator of the reinforcing effects of addictive substances. To investigate the possible role of increased dopamine availability in the pathogenesis of HIV dementia, simian immunodeficiency virus (SIV)-infected monkeys were treated with dopaminergic drugs (selegiline or L-DOPA). Both substances increased intracerebral SIV expression, combined with aggravation of infection-related neuropathology and ultrastructural alterations of dendrites in dopaminergic areas (spongiform polioencephalopathy) in asymptomatic animals. Moreover, this treatment resulted in enhanced TNF-alpha expression in the brains of SIV-infected animals. These findings indicate a synergistic interaction between dopamine and SIV infection on microglia activation, leading to increased viral replication and production of neurotoxic substances. Our results suggest that increased dopamine availability through dopaminergic medication or addictive substances may potentiate HIV dementia.

Measles virus interactions with cellular receptors: consequences for viral pathogenesis.

Although CNS complications occurring early and late after acute measles are a serious problem and often fatal, the transient immunosuppression lasting for several weeks after the rash is the major cause of measles-related morbidity and mortality worldwide. This review is focused on the interactions of measles virus (MV) with cellular receptors on neural and lymphoid cells which are important elements in viral pathogenesis. First, the cognate MV receptors, CD46 and CD150, are important components of viral tropism by mediating binding and entry. Second, however, additional unknown cellular surface molecules may (independently of viral uptake) after interaction with the MV glycoprotein complex act as signaling molecules and thereby modulate cellular survival, proliferation, and specific functions.

Disruption of Akt kinase activation is important for immunosuppression induced by measles virus.

Surface-contact-mediated signaling induced by the measles virus (MV) fusion and hemagglutinin glycoproteins is necessary and sufficient to induce T-cell unresponsiveness in vitro and in vivo. To define the intracellular pathways involved, we analyzed interleukin (IL)-2R signaling in primary human T cells and in Kit-225 cells. Unlike IL-2-dependent activation of JAK/STAT pathways, activation of Akt kinase was impaired after MV contact both in vitro and in vivo. MV interference with Akt activation was important for immunosuppression, as expression of a catalytically active Akt prevented negative signaling by the MV glycoproteins. Thus, we show here that MV exploits a novel strategy to interfere with T-cell activation during immunosuppression.

Measles virus-induced promotion of dendritic cell maturation by soluble mediators does not overcome the immunosuppressive activity of viral glycoproteins on the cell surface.

Measles virus (MV) infection promotes maturation of dendritic cells (DC), but also interferes with DC functions, and MV renders the DC inhibitory for T cell proliferation. We now describe that MV infection triggers the release of type I IFN from monocyte-derived DC (Mo-DC) which contributes to DC maturation. There is no evidence that soluble mediators are released interfering with the stimulatory activity of uninfected DC. Since inhibition of allogeneic T cell proliferation was unaffected by a fusion inhibitory peptide (Z-fFG), MV infection of T cells did not contribute to inhibition. Allogeneic T cell proliferation depended on the percentage of DC expressing MV F/H glycoproteins within the DC population and their surface expression levels, was induced upon addition of UV-inactivated MV to a mixed lymphocyte reaction stimulated by lipopolysaccharide-matured DC, and was not induced by DC infected with a recombinant MV encoding the ectodomain of vesicular stomatitis virus G protein (MG/FV) instead of the MV glycoproteins. Similarly, DC infected with MV, but not with MG/FV inhibited mitogen-induced proliferation of T cells. Thus, a dominant inhibitory signal is delivered to T cells by the MV glycoproteins on the surface of DC overcoming positive signals by co-stimulatory molecules promoted by maturation factors released from infected DC.

Relationship between viral load in blood, cerebrospinal fluid, brain tissue and isolated microglia with neurological disease in macaques infected with different strains of SIV.

The role of the viral burden in the brain for the pathogenesis of human immunodeficiency virus-associated neurological disorders is still unclear. To address this issue, we have quantified the viral load in plasma, cerebrospinal fluid (CSF) and brain tissue of macaques infected with simian immunodeficiency virus (SIV). We discovered that the viral strain used for infection determines the replicative capacity in microglial cells as well as the extent of neuropathological lesions and the occurrence of neurological symptoms. Moreover, the viral load in the brain parenchyma correlated with the development of overt neurological disease whereas the one in plasma did not. By comparing the viral load in three different compartments, we demonstrated that the viral burden in the CSF is influenced both by the viral replication in the periphery as well as in the brain parenchyma. According to these results, it is not the absolute amount of viral load in the CSF but rather the viral antigen contributed by the viral production within the brain which correlates with the development of neurological disease. In longitudinal studies, we observed that this autochthonous virus production, as evidenced by a ratio of the viral load in CSF to the one in plasma, takes place for a prolonged period of time before overt neurological signs are manifested. This finding suggests that this ratio could be used as a prognostic marker for immunodeficiency virus-induced neurological disease.

Antibodies to CD9, a tetraspan transmembrane protein, inhibit canine distemper virus-induced cell-cell fusion but not virus-cell fusion.

Canine distemper virus (CDV) causes a life-threatening disease in several carnivores including domestic dogs. Recently, we identified a molecule, CD9, a member of the tetraspan transmembrane protein family, which facilitates, and antibodies to which inhibit, the infection of tissue culture cells with CDV (strain Onderstepoort). Here we describe that an anti-CD9 monoclonal antibody (MAb K41) did not interfere with binding of CDV to cells and uptake of virus. In addition, in single-step growth experiments, MAb K41 did not induce differences in the levels of viral mRNA and proteins. However, the virus release of syncytium-forming strains of CDV, the virus-induced cell-cell fusion in lytically infected cultures, and the cell-cell fusion of uninfected with persistently CDV-infected HeLa cells were strongly inhibited by MAb K41. These data indicate that anti-CD9 antibodies selectively block virus-induced cell-cell fusion, whereas virus-cell fusion is not affected.

Measles virus-induced immunosuppression in vitro is independent of complex glycosylation of viral glycoproteins and of hemifusion.

Expression of the measles virus (MV) F/H complex on the surface of viral particles, infected cells, or cells transfected to express these proteins (presenter cells [PC]) is necessary and sufficient to induce proliferative arrest in both human and rodent lymphoid cells (responder cells [RC]). This inhibition was found to occur independent of apoptosis and soluble mediators excluded by a pore size filter of 200 nm released from either PC or RC. We now show that reactive oxygen intermediates which might be released by RC or PC also do not contribute to MV-induced immunosuppression in vitro. Using an inhibitor of Golgi-resident mannosidases (deoxymannojirimycin), we found that complex glycosylation of the F and H proteins is not required for the induction of proliferative arrest of RC. As revealed by our previous studies, proteolytic cleavage of the MV F protein precursor into its F1 and F2 subunits, but not of F/H-mediated cellular fusion, was found to be required, since fusion-inhibitory peptides such as Z-D-Phe-L-Phe-Gly (Z-fFG) did not interfere with the induction of proliferative inhibition. We now show that Z-fFG inhibits cellular fusion at the stage of hemifusion by preventing lipid mixing of the outer membrane layer. These results provide strong evidence for a receptor-mediated signal elicited by the MV F/H complex which can be uncoupled from its fusogenic activity is required for the induction of proliferative arrest of human lymphocytes.

DNA vaccination with both the haemagglutinin and fusion proteins but not the nucleocapsid protein protects against experimental measles virus infection.

Plasmids that expressed the nucleocapsid, haemagglutinin and fusion proteins of measles virus (MV) were used to immunize cotton rats (Sigmodon hispidus) against intranasal MV infection. After immunization with all three plasmids, T cell responses and MV-specific antibodies were induced. A reduction in virus titre was observed in lung tissue from animals immunized with plasmids expressing the viral glycoproteins. Histologically, however, a moderate peribronchitis was observed after immunization with the plasmid expressing the fusion protein whereas, after immunization with plasmids expressing haemagglutinin or both glycoproteins, only mild or focal peribronchitis was seen. Immunization with the nucleocapsid did not reduce virus titres, probably because of the failure to induce neutralizing antibodies. A disadvantage of plasmid immunization was its inefficacy in the presence of MV-specific 'maternal' antibodies. This indicates that genetic immunization has to be improved to be a useful alternative vaccine against measles.

Abundant defective viral particles budding from microglia in the course of retroviral spongiform encephalopathy.

A pathogenetic hallmark of retroviral neurodegeneration is the affinity of neurovirulent retroviruses for microglia cells, while degenerating neurons are excluded from retroviral infections. Microglia isolated ex vivo from rats peripherally infected with a neurovirulent retrovirus released abundant mature type C virions; however, infectivity associated with microglia was very low. In microglia, viral transcription was unaffected but envelope proteins were insufficiently cleaved into mature viral proteins and were not detected on the microglia cell surface. These microglia-specific defects in envelope protein translocation and processing not only may have prevented formation of infectious virus particles but also may have caused further cellular defects in microglia with the consequence of indirect neuronal damage. It is conceivable that similar events play a role in neuro-AIDS.

Proteolytic cleavage of the fusion protein but not membrane fusion is required for measles virus-induced immunosuppression in vitro.

Immunosuppression induced by measles virus (MV) is associated with unresponsiveness of peripheral blood lymphocytes (PBL) to mitogenic stimulation ex vivo and in vitro. In mixed lymphocyte cultures and in an experimental animal model, the expression of the MV glycoproteins on the surface of UV-inactivated MV particles, MV-infected cells, or cells transfected to coexpress the MV fusion (F) and the hemagglutinin (H) proteins was found to be necessary and sufficient for this phenomenon. We now show that MV fusion-inhibitory peptides do not interfere with the induction of immunosuppression in vitro, indicating that MV F-H-mediated fusion is essentially not involved in this process. Proteolytic cleavage of MV F(0) protein by cellular proteases, such as furin, into the F(1)-F(2) subunits is, however, an absolute requirement, since (i) the inhibitory activity of MV-infected BJAB cells was significantly impaired in the presence of a furin-inhibitory peptide and (ii) cells expressing or viruses containing uncleaved F(0) proteins revealed a strongly reduced inhibitory activity which was improved following trypsin treatment. The low inhibitory activity of effector structures containing mainly F(0) proteins was not due to an impaired F(0)-H interaction, since both surface expression and cocapping efficiencies were similar to those found with the authentic MV F and H proteins. These results indicate that the fusogenic activity of the MV F-H complexes can be uncoupled from their immunosuppressive activity and that the immunosuppressive domains of these proteins are exposed only after proteolytic activation of the MV F(0) protein.

Dopamine activates HIV in chronically infected T lymphoblasts.

HIV infection is associated with a marked vulnerability of the dopaminergic system. We found recently that dopaminergic substances increase brain pathology in the simian model of HIV infection. In the current study we used the chronically HIV-infected T-lymphoblasts ACH-2 to elucidate the effects of dopamine (DA) on HIV infection. Cells were exposed to various concentrations of DA for 24 hours. Flow cytometry measurements demonstrated that DA induced a concentration-dependent HIV activation. To study the mechanism of action of DA, cells were treated besides DA with glutathione, one of the main components of cellular defense mechanisms against oxidative stress as well as its indirect precursor N-acetylcysteine. Treatment with these antioxidants attenuated DA-induced-HIV activation indicating that changes in cellular redox states might have been the causative factor for the observed effect. Our data suggest that HIV activation is tightly linked to intracellular oxidant/antioxidant levels and that excessive DA exposure may modulate cellular vulnerability to HIV.

Effects of Selegiline in a retroviral rat model for neurodegenerative disease.

Upon inoculation into neonatal rats, murine leukemia virus (MuLV) NT40 causes a non-inflammatory degeneration of the central nervous system. While microglia cells appear to be the major target cells within the brain parenchyma for neurovirulent MuLV, degenerating neurons do not express retroviral gene products. In order to protect rats from neuronal damage we treated retrovirally infected rats once with monoamine oxidase (MAO) B inhibitor Selegiline which--under different conditions--exerts neuroprotective effects. Unexpectedly, when administered at 17 days post-infection (d.p.i.) a single intraperitoneal dose of Selegilin (1 mg/kg bodyweight) significantly shortened the incubation period for neurological disease. In contrast, Selegiline given in a lower dosage (0.05 mg/kg bodyweight) and/or at a different time point (13 d.p.i.) at the low (0.05 mg/kg bodyweight) and the high dose (1.0 mg/kg bodyweight) had no effect on the outcome of neurological disease. Animals treated with Selegiline (1.0 mg/kg bodyweight at 17 d.p.i.) contained higher amounts of viral loads in the CNS, higher numbers of brain cells expressing major histocompatibility complex class II molecules, and exhibited inhibition of MAO-B in comparison to untreated yet infected (control) animals. Supposedly, Selegiline activated the major target cell population of the CNS for MuLV-NT40, microglia, with the consequence of enhanced susceptibility for retroviral infection and triggered endogenous mechanism(s) involved in the pathogenesis of retroviral neurodegeneration.

A recombinant measles vaccine virus expressing wild-type glycoproteins: consequences for viral spread and cell tropism.

Wild-type, lymphotropic strains of measles virus (MV) and tissue culture-adapted MV vaccine strains possess different cell tropisms. This observation has led to attempts to identify the viral receptors and to characterize the functions of the MV glycoproteins. We have functionally analyzed the interactions of MV hemagglutinin (H) and fusion (F) proteins of vaccine (Edmonston) and wild-type (WTF) strains in different combinations in transfected cells. Cell-cell fusion occurs when both Edmonston F and H proteins are expressed in HeLa or Vero cells. The expression of WTF glycoproteins in HeLa cells did not result in syncytia, yet they fused efficiently with cells of lymphocytic origin. To further investigate the role of the MV glycoproteins in virus cell entry and also the role of other viral proteins in cell tropism, we generated recombinant vaccine MVs containing one or both glycoproteins from WTF. These viruses were viable and grew similarly in lymphocytic cells. Recombinant viruses expressing the WTFH protein showed a restricted spread in HeLa cells but spread efficiently in Vero cells. Parental WTF remained restricted in both cell types. Therefore, not only differential receptor usage but also other cell-specific factors are important in determining MV cell tropism.

Expression of measles virus V protein is associated with pathogenicity and control of viral RNA synthesis.

Nonstructural proteins encoded by measles virus (MV) include the V protein which is translated from an edited P mRNA. V protein is not associated with intracellular or released viral particles and has recently been found to be dispensable for MV propagation in cell culture (H. Schneider, K. Kaelin, and M. A. Billeter, Virology 227:314-322, 1997). Using recombinant MVs (strain Edmonston [ED]) genetically engineered to overexpress V protein (ED-V+) or to be deficient for V protein (ED-V-), we found that in the absence of V both MV-specific proteins and RNAs accumulated to levels higher than those in the parental MV molecular clone (ED-tag), whereas MV-specific gene expression was strongly attenuated in human U-87 glioblastomas cells after infection with ED-V+. The titers of virus released from these cells 48 h after infection with either V mutant virus were lower than those from cells infected with ED-tag. Similarly, significantly reduced titers of infectious virus were reisolated from lung tissue of cotton rats (Sigmodon hispidus) after intranasal infection with both editing mutants compared to titers isolated from ED-tag-infected animals. In cell culture, expression of V protein led to a redistribution of MV N protein in doubly transfected Cos-7 cells, indicating that these proteins form heterologous complexes. This interaction was further confirmed by using a two-hybrid approach with both proteins expressed as Gal4 or VP16 fusion products. Moreover, V protein efficiently competed complexes formed between MV N and P proteins. These findings indicate that V protein acts to balance accumulation of viral gene products in cell culture, and this may be dependent on its interaction with MV N protein. Furthermore, expression of V protein may contribute to viral pathogenicity in vivo.

Differential receptor usage by measles virus strains.

Recently, we demonstrated that infection of cells with all measles virus (MV) strains tested was inhibited by antibodies against CD46, although not all strains caused downregulation of the MV receptor CD46 from the surface of human cells. We now show that infection of cells with MV strain WTFb, a variant of wild-type isolate WTF which has been isolated and propagated on human BJAB cells, is not inhibited by antibodies against CD46. In contrast, infection of cells with the closely related strain WTFv, a Vero cell-adapted variant of WTF, is inhibited by antibodies against CD46. This observation led us to investigate the interaction of these viruses and the vaccine strain Edmonston (Edm) with CD46 and target cells. Cellular receptors with high affinity binding for WTFb are present on BJAB cells, but not on transfected CD46-expressing CHO cells. In contrast to the Edm strain, virus particles and solubilized envelope glycoproteins of WTFb have a very limited binding capacity to CD46. Furthermore, we show that recombinant soluble CD46 either does not bind, or binds very weakly, to WTFb glycoproteins expressed on the cell surface. Our findings indicate that wild-type MV strain WTFb and vaccine strain Edm use different binding sites on human cells. In addition, the results suggest that MV strains may alternatively use CD46 and an unknown molecule as receptors, and that the degree of usage of both receptors may be MV strain-specific.

Oral or parenteral administration of replication-deficient adenoviruses expressing the measles virus haemagglutinin and fusion proteins: protective immune responses in rodents.

The genes encoding the measles virus (MV) haemagglutinin (H) and fusion (F) proteins were placed under the control of the human cytomegalovirus immediate early promoter in a replication-deficient adenovirus vector. Immunofluorescence and radioimmune precipitation demonstrated the synthesis of each protein and biological activity was confirmed by the detection of haemadsorption and fusion activities in infected cells. Oral as well as parenteral administration of the H-expressing recombinant adenovirus elicited a significant protective response in mice challenged with MV. While the F-expressing adenovirus failed to protect mice, cotton rats immunized with either the H- or F-expressing recombinant showed reduced MV replication in the lungs. Antibodies elicited in mice following immunization with either recombinant had no in vitro neutralizing activity, suggesting a protective mechanism involving a cell-mediated immune response. This study demonstrates the feasibility of using oral administration of adenovirus recombinants to induce protective responses to heterologous proteins.

Absence of measles virus receptor (CD46) in lesions of subacute sclerosing panencephalitis brains.

In this study we investigated pathological changes of the expression of the measles virus (MV) receptor, CD46, in subacute sclerosing panencephalitis (SSPE) brains. We analyzed CD46 expression in lesions of brain specimens from five SSPE patients in comparison to uninfected regions of the same brains and to normal human brains. The correlation between CD46 and MV infection, in individual cells in SSPE brains, was analyzed by double-staining procedures using monoclonal antibodies (mAbs) and in situ hybridization to detect MV-specific mRNAs. We found that CD46 was expressed at relatively low levels by neurons and astrocytes in normal brains in comparison to neuroblastoma and astrocytoma cell lines. Within heavily infected (MV-positive) brain lesions of all five SSPE cases, CD46 was either not detected or was expressed to a lesser degree by neural cells, irrespective of whether MV antigens were detectable or not. In contrast, normal levels of CD46 were found in SSPE brain tissue distant from the lesion. Using in situ hybridization, mRNAs of both MV nucleocapsid and MV hemagglutinin (MV-H) were detected in all SSPE lesions, while no or only small amounts of MV-H protein were detected. MV-infected neurons were never found to express CD46. Although a strict correlation between levels of the MV-H protein and the absence CD46 could not be seen, these findings suggest that the CD46 expression is reduced by the MV infection in lesions of SSPE brains.