Rabies virus selectively alters 5-HT1 receptor subtypes in rat brain

Rabies virus infection in man induces a series of clinical symptoms, some suggesting involvement of the central serotonergic system. The results of the present study show that, 5 days after rabies virus infection in rat, the total reversible high-affinity binding of [3H]5-HT in the hippocampus is not affected, suggesting that 5-HT1A binding is not altered. 5-HT1B sites identified by [125I]cyanopindolol binding are not affected in the cortex 3 and 5 days after the infection. Accordingly, the cellular inhibitory effect of trifluoromethylphenylpiperazine (TFMPP) on the [3H]acetylcholine-evoked release, presumably related to 5-HT1B receptor activity, is not modified 3 days after infection. In contrast, [3H]5-HT binding determined in the presence of drugs masking 5-HT1A, 5-HT1B and 5-HT1C receptors, is markedly (50%) reduced 3 days after the viral infection. These results suggest that 5-HT1D-like receptor subtypes may be affected specifically and at an early stage after rabies viral infection.

Rabies RNA synthesis, detected with cDNA probes, as a marker for virus transport in the rat nervous system

The kinetics of viral RNA synthesis in different parts of the rat brain, infected with fixed or street rabies virus strains, is correlated with their anatomical neuronal connections with the masseter muscles, using hybridization with rabies cDNA probes. Viral RNA synthesis is first detected in the brain-stem and in the pons where the direct anatomical projection of the masseter muscle nervous arborization into the sensory and motor nuclei is located, through the trigeminus nerve. Rabies RNA detection is delayed in the other regions of the rat brain depending on the time course of virus transport from the trigeminal nuclei through multiple nervous connections.

Inhibition of rabies virus infection in cultured rat cortical neurons by an N-methyl-D-aspartate noncompetitive antagonist, MK-801

A noncompetitive N-methyl-D-aspartate (NMDA) antagonist, MK-801 (0.5 to 2.0 mM), inhibits rabies virus infection in rat primary cortical neurons, whereas the competitive NMDA antagonist AP5 has no effect. The results suggest that MK-801-mediated inhibition of rabies virus replication, although selective, is not operating through the high-affinity binding site mechanism.

Polymerase chain reaction amplification of rabies virus nucleic acids from total mouse brain RNA

In an attempt to improve the sensitivity of the rabies genome hybridization test, PCR amplification was used following reverse transcription of rabies RNA extracted from infected brain. Presence of amplified DNA is demonstrated with either cDNA synthesized from the antigenomic primer or from antimessenger primer.

Continuous delivery of colchicine in the rat brain with osmotic pumps for inhibition of rabies virus transport

Rabies virus is a neurotropic agent which spreads in the CNS via axonal transport. Previous studies had shown that this axonal transport through the brain could be inhibited by stereotaxic administration of colchicine; however, this inhibition was reversible. We describe here a method to enhance the duration of this colchicine-mediated inhibition by delivering the drug continuously in the rat brain with osmotic pumps.

The anterograde transport of rabies virus in rat sensory dorsal root ganglia neurons

We have previously described the capacity of neurites extending from cultured rat sensory dorsal root ganglia (DRG) neurons to transport rabies virus through axoplasm in the retrograde direction. Here we report the infection of cultured neurons derived from the DRG and the subsequent anterograde transport of rabies virus from the infected cell somas through the extending neurites to its release into the culture supernatant. Viral transport was monitored by titration of the virus yield in the external compartment. Both early and late transport mechanisms of rabies virions were identified. The first one occurred a few hours post-infection and was undetectable 6 h later, before the initiation of viral replication. The velocity of this first wave of infective virions was in the range of 100 to 400 mm/day. The early viral transport was probably the result of a direct translocation of infective virions from the somatic site of entry to the neuritic extensions and subsequent release into the culture medium without replication in the cellular perikaryon. The second virus transport peak was detected 48 h post-infection. In this case, the virions detected in the neuritic compartment were presumably the progeny of the inoculated virus which had replicated in the perikaryon before the viral transport occurs. Using a four-compartment culture device we were able to demonstrate, simultaneously, retrograde and anterograde transport of the virus. The presence of antirabies serum in contact with the exposed neurites did not inhibit either the retrograde or the anterograde transport mechanisms. The viral release from the neuritic extensions after the fast anterograde transport was evaluated to be in the range of 150 to 300 infectious virions per bundle of neurites per day.

Completion of the rabies virus genome sequence determination: highly conserved domains among the L (polymerase) proteins of unsegmented negative-strand RNA viruses

We have now completed the rabies genome structure by the cloning and the sequencing of the entire L gene and the 5' untranscribed region. The L gene encodes a single open reading frame 2142 amino acids in length (244,206 Da) that corresponds to the viral RNA-dependent RNA polymerase. In contrast with other isofunctional proteins, the rabies polymerase exhibits a high degree of homology with the vesicular stomatitis virus polymerase, and a lesser degree, although significant, with those of Sendai virus and Newcastle disease virus, which suggests a differential evolution of the different cistrons. We have observed several strongly conserved stretches which may designate the independent functional domains of this multifunctional protein. In addition to the conservation of related transcription signals (N. Tordo et al. (1986) Proc. Natl. Acad. Sci. USA 83, 3914-3918.), this highlights the striking selective pressure on elements involved in transcription and replication mechanisms, and provides further evidence for a common ancestry of Rhabdoviridae and Paramyxoviridae families. The terminal complementarity observed in the rabies genome suggests the conservation of important genomic signals.

Rapid diagnosis of rabies infection by means of a dot hybridization assay

Dot hybridization was used to detect specific rabies RNA in brains, either from experimental infection in mouse or from brain material to be processed for routine diagnosis. 32P cDNA probes were employed to identify minute amounts of specific viral RNA. Purified RNA was obtained after phenol extraction. The RNA was fixed on nylon membranes and hybridized with a pool of M13 inserts complementary to 200-400 nucleotides of each rabies gene and mRNA. Hybridized, labelled probes were detected by autoradiography. There was strong cross-hybridization between fixed rabies and street rabies virus RNA, which enable the detection of field strains for diagnosis purpose using a fixed rabies (PV strain) cDNA. A positive response was obtained with as little as 80 ng of brain RNA material from a fixed rabies-infected mouse. Detection of viral RNA was still specific 1 week after death, the brain material being left at room temperature. A total correlation was found when the samples were examined in parallel using a fluorescent rabies-specific antibody and by virus isolation on murine neuroblastoma cells. These data show that the use of rabies-specific cDNA probes in a dot-blot hybridization assay has great potential for the diagnosis of rabies.

Walking along the rabies genome: is the large G-L intergenic region a remnant gene?

Rabies cDNA clones, obtained by "walking along the genome" using two successive DNA primers, have allowed the sequence determination of the genes encoding the N, M1, M2, G, and the beginning of the L protein as well as the rabies intergenic regions. Start and stop transcription signals located at the border of each gene encoding a protein have been identified and are similar to the corresponding signals from vesicular stomatitis virus (VSV) and Sendai virus. Except for limited stretches of the nucleoprotein, there is no homology between corresponding structural proteins of these three viruses. Rabies intergenic regions are variable both in length and sequence. Evidence for the existence of a remnant protein gene in the 423 nucleotide long G-L intergenic region is presented. This finding is discussed in terms of the evolution of unsegmented negative-strand RNA viruses.

Primary structure of leader RNA and nucleoprotein genes of the rabies genome: segmented homology with VSV

We have determined the nucleotide sequence of the 3'region of the rabies genome (PV strain). This work is a first step in a project aimed at establishing the complete primary structure. From the 3'nucleotide sequence of the RNA genome, an octadecanucleotide complementary to the 3'extremity was constructed and used to prime cDNA synthesis. Two overlapping recombinant cDNA clones hybridizing with the nucleoprotein mRNA (NmRNA) were isolated and sequenced. The 1500 first nucleotides of the rabies genome cover two transcriptional units: the leader RNA and the NmRNA which was shown to be initiated around residue 59 by S1 nuclease protection experiments. Comparison between rabies PV and CVS strains up to residue 180 suggests a rapid evolution in the leader region. Studies of the sequence relationships between the 3'regions of two Rhabdoviruses, rabies virus and Vesicular Stomatitis Virus (VSV), demonstrate that there is a segmented homology. Stretches of highly conserved amino acids possibly involved in the interaction with the RNA genome were observed in the N protein, despite a wide divergence in the remaining sequence. In addition, the high homology between the transcription start and stop signals reflects the conservation of a similar transcriptional mechanism in these two non segmented negative strand RNA viruses.

Protein synthesis in VSV infected CNS, neuroblastoma and BHK cell lines

The protein synthesis of VSV infected CNS of mice was analysed by SDS-PAGE. Cellular and viral protein synthesis in the CNS were also compared to VSV infected neuroblastoma cells (clone NS 20) and fibroblasts (BHK21). Inhibition of host protein synthesis was observed in the three systems tested. However, this inhibition was shown to occur faster in neuroblastoma cells than in BHK cells, whereas it proceeded progressively in the brain. Thus the shut off of host cell protein synthesis by VSV seems to be a general phenomenon that occurs in vivo as well as in vitro. VSV protein from the CNS, or from neuroblastoma cells and fibroblasts were found to migrate similarly in SDS-PAGE. The viral L protein synthesis was found to be particularly active in the CNS, with respect to that observed in NS20 and BHK cells. The viral glycoprotein failed to be detected in the VSV infected mouse brain in our experimental conditions. The results show that VSV infection in vivo occurs with some difference with that of non neuronal cells, and that in vivo studies suggest the existence of cellular modulation that should be taken in account in the pathogenesis of this rhabdovirus.

[Molecular and cellular aspects of the pathogenesis of rabies]

The true causes of the beginning of the symptoms and the eventual mortality of rabies are not very well known. Immunopathological factors certainly intervene, but they only complete a process of neuronal alteration. Contrary to other viruses which lyse the infected cells, the rabies virus does not necessarily destroy the neurones which are the preferential target cells. On the contrary, RNA synthesis is stimulated during CNS infection by the rabies virus. Synthesis is not stimulated by the multiplication of the rabies virus in the cell lines of fibroblastic or neuronal origin, which suggests the existence of regulations on neuronal level functions in the CNS. Nevertheless, the protein virus synthesis seemed equally submitted to specific neuron regulations or neuronal populations. The traces of a neuronal dysfunction during infection by the rabies virus have been investigated, using as a probe, variations of muscarinic acetyl choline receptors in the CNS. The presence of these receptors is revealed by the use of an antagonistic ligand of acetyl choline, quinuclidinyl benzylate. We were able to show that as regards the rat, the appearance of the first signs of paralysis is correlated with the diminution of the receptor binding. Thus, it seems that the presence of the virus in the CNS provokes neuronal alteration which can in itself condition the viral multiplication and eventually modulate the viral pathogenesis expression.

[Poly(A) sequences associated with rabies virus mRNA (author's transl)]

Polyadenylate sequences have been found in numerous RNA with messenger function, and in several viruses in which the genome is its own messenger. With rabies virus we observed a fixation of viral RNA on Sepharose poly(U) column in the mRNA region (8-25S) and in the double stranded RNA region (25-35S). These percentages of fixation are very low compared to those of VSV in the same regions, but they are significant compared to the uninfected cells and to Lumbo RNA fixation. In order to determine the nature of the RNA bound to poly(U)-sepharose, fractions of the region 8-25S and 25-35S obtained after elution with formamide were centrifuged through a sucrose gradient. We found in the 8-25S region an RNA profile which was identical to the initial profile of this region. In contrast, the 25-35S region exhibited a heterogenous profile which looked like the total virus RNA profile obtained from infected cells. This may be due to the presence of double-stranded RNA in this region which is partially adsorbed on poly(U) and denatured by formamide in RNA molecules of genome and mRNA size. We conclude from our results that the low percentage of fixation of rabies mRNA on poly(U)-sepharose columns could mean either that a small number of molecules are polyadenylated or that a large number exhibit poly(A) tracts too short to be retained by the poly(U).

Rna syntheses in BHK21 cells infected by rabies virus

In BHK21 cells infected by rabies virus, viral RNA syntheses are detectable from the first 4 hours till at least 20h after the adsorption period. Cellular RNA syntheses are not inhibited by the virus. The viral RNA syntheses are 10 times lower than syntheses induced in infected cells by VSV. They are also slower since the maximum is between 8 and 12h after infection. At least 3 categories of molecules are synthesized: 1) short (+) molecules sedimenting between 8 and 25 S, and possibly at 30 S; 2) partially or totally double-stranded structures sedimenting between 25 and 35 S; 3) (+) and (--) molecules of the genome length. The relative amount of these 3 categories of molecules does not seem to vary during the viral cycle.