Cross-reactive antigenicity of nucleoproteins of lyssaviruses recognized by a monospecific antirabies virus nucleoprotein antiserum on paraffin sections of formalin-fixed tissues

Incidence and seroprevalence of rabies virus in humans, dogs and other animal species in Africa, a systematic review and meta-analysis

Rabies is transmitted to humans mainly by dogs but also by other animal species. Reliable data on the incidence of Rabies virus (RABV) in humans, dogs, and other animal species in Africa, could be essential in the implementation of a global strategic plan to eliminate the RABV by 2030 as adopted by the WHO, OIE, and FAO. We searched the Pubmed, Embase, Scopus, African Journal Online, and African Index Medicus databases for relevant studies that report data on the incidence of RABV in Africa up to February 17, 2020. Information on active and past RABV exposures in various categories of dogs, humans and other animal species were extracted. Incidence and seroprevalence estimates were pooled using a random-effect meta-analysis. We included 73 articles which provided 142 RABV incidence and seroprevalence records in 21 African countries. The estimated incidence of RABV in 222 humans, 15,600 dogs, and 12,865 other animal species was 83.4% (95% CI = 64.6–96.5), 44.1% (95% CI = 35.1–53.4), and 41.4% (95% CI = 29.6–53.8), respectively. The estimated seroprevalence of RABV in 420 humans, 3577 dogs, and 8,55 other animal species was 33.8% (95% CI = 21.9–46.8), 19.8% (95% CI = 13.3–27.3), and 3.6% (95% CI = 0.3–9.2), respectively. The incidence of RABV in general was higher in suspected rabid dogs, other animal species of the Orders Perissodactyla, Artiodactyla and Carnivora. The incidence of RABV was higher for humans in regions of West and East Africa, for dogs in urban areas and in regions of Central and South Africa, and for animals of the order Perissodactyla in urban areas. This meta-analysis demonstrated a high incidence of RABV in Africa. Itis necessary to improve surveillance system to provide reliable data on RABV in Africa, essential for the implementation of an effective control strategy.

Morphological features of large layer V pyramidal neurons in cortical motor-related areas of macaque monkeys: analysis of basal dendrites

In primates, large layer V pyramidal neurons located in the frontal motor-related areas send a variety of motor commands to the spinal cord, giving rise to the corticospinal tract, for execution of skilled motor behavior. However, little is known about the morphological diversity of such pyramidal neurons among the areas. Here we show that the structure of basal dendrites of the large layer V pyramidal neurons in the dorsal premotor cortex (PMd) is different from those in the other areas, including the primary motor cortex, the supplementary motor area, and the ventral premotor cortex. In the PMd, not only the complexity (arborization) of basal dendrites, i.e., total dendritic length and branching number, was poorly developed, but also the density of dendritic spines was so low, as compared to the other motor-related areas. Regarding the distribution of the three dendritic spine types identified, we found that thin-type (more immature) spines were prominent in the PMd in comparison with stubby- and mushroom-type (more mature) spines, while both thin- and stubby-type spines were in the other areas. The differential morphological features of basal dendrites might reflect distinct patterns of motor information processing within the large layer V pyramidal neurons in individual motor-related areas.

Segmentation of the rabies virus genome

We established a system for the recovery of a segmented recombinant rabies virus, the virus genome RNA of which was divided into two parts: segment 1 encoding the nucleoprotein, phosphoprotein, matrix protein, and glycoprotein genes, and segment 2 encoding the large RNA-dependent RNA polymerase gene. The morphology of the segmented recombinant rabies virus was bullet-like in shape with a length of approximately 130 nm, which is shorter than the 200-nm long non-segmented recombinant rabies virus. The segmented recombinant rabies virus was maintained for at least 18 passages. The virus multiplication rate of the segmented recombinant rabies virus was lower than that of the non-segmented recombinant rabies virus during the passages, and the relative amounts of virus genome RNAs for segment 1 and segment 2 differed in the supernatant of the segmented recombinant rabies virus infected cells. These results suggest that the segmented recombinant rabies virus packages either segment 1 or segment 2 into each virus particle. Thus, co-infection with segmented recombinant rabies virus particles packaging segment 1 or segment 2 may be necessary for the production of progeny virus.

Increased transgene expression level of rabies virus vector for transsynaptic tracing

Viral vectors that can infect neurons transsynaptically and can strongly express foreign genes are useful for investigating the organization of neural circuits. We previously developed a propagation-competent rabies virus (RV) vector based on a highly attenuated HEP-Flury strain (rHEP5.0-CVSG), which selectively infects neurons and propagates between synaptically connected neurons in a retrograde direction. Its relatively low level of transgene expression, however, makes immunostaining necessary to visualize the morphological features of infected neurons. To increase the transgene expression level of this RV vector, in this study we focused on two viral proteins: the large protein (L) and matrix protein (M). We first attempted to enhance the expression of L, which is a viral RNA polymerase, by deleting the extra transcription unit and shortening the intergenic region between the G and L genes. This viral vector (rHEP5.0-GctL) showed increased transgene expression level with efficient transsynaptic transport. We next constructed an RV vector with a rearranged gene order (rHEP5.0-GML) with the aim to suppress the expression of M, which plays a regulatory role in virus RNA synthesis. Although this vector showed high transgene expression level, the efficiency of transsynaptic transport was low. To further evaluate the usability of rHEP5.0-GctL as a transsynaptic tracer, we inserted a fluorescent timer as a transgene, which changes the color of its fluorescence from blue to red over time. This viral vector enabled us the differentiation of primary infected neurons from secondary infected neurons in terms of the fluorescence wavelength. We expect this propagation-competent RV vector to be useful for elucidating the complex organization of the central nervous system.

Association between RABV G Proteins Transported from the Perinuclear Space to the Cell Surface Membrane and N-Glycosylation of the Sequon Asn(204)

In this study, G proteins of the rabies virus (RABV) Kyoto strain were detected in the cytoplasm but not distributed at the cell membrane of mouse neuroblastoma (MNA) cells. G proteins of CVS-26 were detected in both the cell membrane and perinuclear space of MNA cells. We found that N-glycosylation of street RABV G protein by the insertion of the sequon Asn(204) induced the transfer of RABV G proteins to the cell surface membrane. Fixed RABV budding from the plasma membrane has been found to depend not only on G protein but also on other structural proteins such as M protein. However, the differing N-glycosylation of G protein could be associated with the distinct budding and antigenic features of RABV in street and fixed viruses. Our study of the association of N-glycan of G protein at Asn(204) with the transport of RABV G protein to the cell surface membrane contributes to the understanding of the evolution of fixed virus from street virus, which in turn would help for determine the mechanism underlying RABV budding and enhanced host immune responses.

Rabies virus vector transgene expression level and cytotoxicity improvement induced by deletion of glycoprotein gene

The glycoprotein (G) of rabies virus (RV) is required for binding to neuronal receptors and for viral entry. G-deleted RV vector is a powerful tool for investigating the organization and function of the neural circuits. It gives the investigator the ability to genetically target initial infection to particular neurons and to control trans-synaptic propagation. In this study we have quantitatively evaluated the effect of G gene deletion on the cytotoxicity and transgene expression level of the RV vector. We compared the characteristics of the propagation-competent RV vector (rHEP5.0-CVSG-mRFP) and the G-deleted RV vector (rHEP5.0-ΔG-mRFP), both of which are based on the attenuated HEP-Flury strain and express monomeric red fluorescent protein (mRFP) as a transgene. rHEP5.0-ΔG-mRFP showed lower cytotoxicity than rHEP5.0-CVSG-mRFP, and within 16 days of infection we found no change in the basic electrophysiological properties of neurons infected with the rHEP5.0-ΔG-mRFP. The mRFP expression level of rHEP5.0-ΔG-mRFP was much higher than that of rHEP5.0-CVSG-mRFP, and 3 days after infection the retrogradely infected neurons were clearly visualized by the expressed fluorescent protein without any staining. This may be due to the low cytotoxicity and/or the presumed change in the polymerase gene (L) expression level of the G-deleted RV vector. Although the mechanisms remains to be clarified, the results of this study indicate that deletion of the G gene greatly improves the usability of the RV vector for studying the organization and function of the neural circuits by decreasing the cytotoxicity and increasing the transgene expression level.

Multisynaptic projections from the ventrolateral prefrontal cortex to hand and mouth representations of the monkey primary motor cortex

Different sectors of the prefrontal cortex have distinct neuronal connections with higher-order sensory areas and/or limbic structures and are related to diverse aspects of cognitive functions, such as visual working memory and reward-based decision-making. Recent studies have revealed that the prefrontal cortex (PF), especially the lateral PF, is also involved in motor control. Hence, different sectors of the PF may contribute to motor behaviors with distinct body parts. To test this hypothesis anatomically, we examined the patterns of multisynaptic projections from the PF to regions of the primary motor cortex (MI) that represent the arm, hand, and mouth, using retrograde transsynaptic transport of rabies virus. Four days after rabies injections into the hand or mouth region, particularly dense neuron labeling was observed in the ventrolateral PF, including the convexity part of ventral area 46. After the rabies injections into the mouth region, another dense cluster of labeled neurons was seen in the orbitofrontal cortex (area 13). By contrast, rabies labeling of PF neurons was rather sparse in the arm-injection cases. The present results suggest that the PF-MI multisynaptic projections may be organized such that the MI hand and mouth regions preferentially receive cognitive information for execution of elaborate motor actions.

Anatomical evidence for the involvement of medial cerebellar output from the interpositus nuclei in cognitive functions

Although the cerebellar interpositus nuclei are known to be involved in cognitive functions, such as associative motor learning, no anatomical evidence has been available for this issue. Here we used retrograde transneuronal transport of rabies virus to identify neurons in the cerebellar nuclei that project via the thalamus to area 46 of the prefrontal cortex of macaques in comparison with the projections to the primary motor cortex (M1). After rabies injections into area 46, many neurons in the restricted region of the posterior interpositus nucleus (PIN) were labeled disynaptically via the thalamus, whereas no neuron labeling was found in the anterior interpositus nucleus (AIN). The distribution of the labeled neurons was dorsoventrally different from that of PIN neurons labeled from the M1. This defines an anatomical substrate for the contribution of medial cerebellar output to cognitive functions. Like the dentate nucleus, the PIN has dual motor and cognitive channels, whereas the AIN has a motor channel only.

Dorsal area 46 is a major target of disynaptic projections from the medial temporal lobe

The medial temporal lobe (MTL) is responsible for various mnemonic functions, such as association/conjunction memory. The lateral prefrontal cortex (LPFC) also plays crucial roles in mnemonic functions and memory-based cognitive behaviors, for example, decision-making. Therefore, it is considered that the MTL and LPFC connect with each other and cooperate for the control of cognitive behaviors. However, there exist very weak, if any, direct inputs from the MTL to the LPFC. Employing retrograde transsynaptic transport of rabies virus, we investigated the organization of disynaptic bottom-up pathways connecting the MTL and the inferotemporal cortex to the LPFC in macaques. Three days after rabies injections into dorsal area 46, a large number of labeled neurons were observed in the MTL, such as the hippocampal formation (including the entorhinal cortex), the perirhinal cortex, and the parahippocampal cortex. In contrast, a majority of the labeled neurons were located in the inferotemporal cortex following rabies injections into ventral area 46 and lateral area 12. Rabies injections into lateral area 9/area 8B labeled only a small number of neurons in the MTL and the inferotemporal cortex. The present results indicate that, among the LPFC, dorsal area 46 is the main target of disynaptic inputs from the MTL.

Segregated pathways carrying frontally derived top-down signals to visual areas MT and V4 in macaques

The bottom-up processing of visual information is strongly influenced by top-down signals, at least part of which is thought to be conveyed from the frontal cortex through the frontal eye field (FEF) and the lateral intraparietal area (LIP). Here we investigated the architecture of multisynaptic pathways from the frontal cortex to the middle temporal area (MT) of the dorsal visual stream and visual area 4 (V4) of the ventral visual stream in macaques. In the first series of experiments, the retrograde trans-synaptic tracer, rabies virus, was injected into MT or V4. Three days after rabies injections, the second-order (disynaptically connected) neuron labeling appeared in the ventral part of area 46 (area 46v), along with the first-order (monosynaptically connected) neuron labeling in FEF and LIP. In the MT-injection case, second-order neurons were also observed in the supplementary eye field (SEF). In the next series of experiments, double injections of two fluorescent dyes, fast blue and diamidino yellow, were made into MT and V4 to examine whether the frontal inputs are mediated by distinct or common neuronal populations. Virtually no double-labeled neurons were observed in FEF or LIP, indicating that separate neuronal populations mediate the frontal inputs to MT and V4. The present results define that the multisynaptic frontal input to V4 arises primarily from area 46v, whereas the input to MT arises from not only area 46v but also SEF, through distinct FEF and LIP neurons. Segregated pathways from the frontal cortex possibly carry the functionally diverse top-down signals to each visual stream.

Differential architecture of multisynaptic geniculo-cortical pathways to V4 and MT

Parallel visual pathways in the primate brain known as the dorsal and ventral streams receive retinal inputs mainly through the magnocellular (M) and parvocellular (P) layers of the lateral geniculate nucleus. Inputs from these layers terminate within distinct parts of layer 4C of V1 (visual area 1). Due to the complexity of M- and P-derived neural connectivity in V1 and higher visual areas, the contributions of M and P inputs to the dorsal and ventral streams remain unclear. Employing retrograde transsynaptic transport of rabies virus, we analyzed the architecture of bottom-up pathways toward ventral stream area V4 (visual area 4) and dorsal stream area MT (middle temporal area). We found that V4 receives both M and P inputs "trisynaptically" from layer 4C via layer 2/3 of V1, whereas MT receives M-dominant input "disynaptically" from layer 4C via layer 4B of V1. V4 also receives disynaptic input from the dorsal stream portion of V2 (visual area 2) (i.e., cytochrome oxidase-stained thick stripes). Moreover, both M and P inputs reach V4 trisynaptically and MT disynaptically through "short-cut" pathways that bypass layer 4C of V1. The differential patterns of multisynaptic geniculo-cortical pathways to V4 and MT imply distinct modes of information processing in the dorsal and ventral streams.

Pathology of the spinal cord of C57BL/6J mice infected with rabies virus (CVS-11 strain)

Fixed rabies viruses (CVS-11 strain) were inoculated intramuscularly to C57BL/6J mice, and the pathomorphological changes of the spinal cord including dorsal root spinal ganglion cells were investigated. At 4 days postinoculation (PI), viral antigens were first detected in the spinal neurons and dorsal root spinal ganglion cells without producing morphological changes. At 5 days PI, mild infiltration of lymphocytes was observed around the central canal, small blood vessels and leptomeninges. Cells positive to anti-Iba1 and anti-GFAP antibodies increased significantly from 3 to 5 days PI, respectively. Microglia changed their morphological forms to be ramified or amoeboid, and astroglia extended their cytoplasm from the leptomeninges to the parenchyma. At 7 days PI, apoptotic cells were found in the spinal cord and dorsal root spinal ganglion using TUNEL. We confirmed that most of T lymphocytes and a minority of microglial cells underwent apoptosis, using a combination of TUNEL and immunostaining with antibodies to viral phosphoprotein, CD3, Iba1 and GFAP. On the other hand, astroglial cells and virus-infected nerve cells were negative against TUNEL and cleaved caspase-3 antibody. These findings indicate that T lymphocytes and microglial cells died by apoptosis, whereas virus-infected nerve cells died by necrosis. This was accompanied by increased numbers and morphological changes of glial cells associated with the pathogenesis of CVS-11 in the C57BL/6J mouse.

A simple and rapid immunochromatographic test kit for rabies diagnosis

In rabies endemic countries, funds and infrastructure are often insufficient to employ the approved gold standard for the definitive diagnosis of rabies: the direct fluorescent test. In the present study, two types (type 1 and 2) of an ICT kit were evaluated for detection of rabies. These were developed using monoclonal antibodies which recognize epitope II and III of the nucleoprotein of rabies virus. Both kits specifically detected all rabies virus strains and there was no cross reactivity with Lyssaviruses (Lagos, Mokola and Duvenhage), Rhabdovirus (VSV and Oita 296/1972) and other common canine‐pathogenic viruses. In type 1, a single type of monoclonal antibody was used. It was capable of detecting recombinant nucleoprotein and showed sensitivity of 95.5% (42/44) and specificity of 88.9% (32/36) using brain samples from rabid dogs. In contrast, type 2 which was made of two different monoclonal antibodies had a lower sensitivity of 93.2% (41/44) and higher specificity of 100% (36/36). These ICT kits provide a simple and rapid method for rabies detection. They need neither cold chain for transportation nor complicated training for personnel. This diagnostic test is suitable for rabies screening, particularly in areas with a high prevalence of rabies and where the fluorescent antibody test is not available.

Topographic distribution of output neurons in cerebellar nuclei and cortex to somatotopic map of primary motor cortex

To investigate the somatotopic organization of the cerebellum, we analysed multisynaptic inputs to the primary motor cortex (MI) using retrograde transneuronal transport of rabies virus. At 3 days after rabies injections into proximal forelimb, distal forelimb and hindlimb representations of the macaque MI, second-order neurons via the thalamus were labeled in the deep cerebellar nuclei, including the dentate (DN), anterior interpositus (AIN) and posterior interpositus nuclei. In the DN, the labeling of both the forelimb and hindlimb was seen mainly in the dorsal aspect. The labeling of the hindlimb was located rostral to that of the forelimb and the labeling of the proximal forelimb was located slightly rostral to that of the distal forelimb. The same rostrocaudal arrangement was observed in the AIN. In the posterior interpositus nucleus, however, labeling from the MI hindlimb and forelimb representations largely overlapped. At the 4-day postinjection period, third-order labeling occurred in Purkinje cells of the cerebellar hemisphere. The Purkinje cell labeling from the forelimb representation, including the proximal and distal regions, was observed primarily in lobules IV-VI and crus I. The proximal forelimb labeling was both rostral and lateral to that of the distal forelimb within lobules IV-VI. However, the hindlimb labeling was seen both rostral and lateral to that of the proximal forelimb within lobules III-VI. These results indicate that the hindlimb, proximal forelimb and distal forelimb are arranged rostrocaudally in the DN and AIN, whereas there is dual somatotopy along the rostrocaudal and lateromedial axes in the cerebellar cortex.

Role of GPI-anchored NCAM-120 in rabies virus infection

Although the neural cell adhesion molecule (NCAM) -140 and -180 have been shown to serve as a receptor for rabies virus (RV), it was not known whether the other major isoform of NCAM, GPI-anchored NCAM-120 functions as RV receptor. In this study, we have established HEp-2 cells stably expressing NCAM-120 or NCAM-140, and their susceptibilities to RV infection were compared. The results demonstrated that NCAM-120 served as virus attachment protein; however, the cells expressing NCAM-120 did not support efficient RV replication. Furthermore, the level of IFN-ss mRNA was apparently elevated in NCAM-120 expressing cells but not in NCAM-140 expressing cells, suggesting that GPI-anchored NCAM-120 suppressed RV replication via induction of IFN-ss even though NCAM-120 was able to promote virus penetration into the cells.

Expression of the nucleoprotein gene of rabies virus for use as a diagnostic reagent

The nucleoprotein (N) gene of rabies virus CTN strain, was cloned, sequenced and expressed in Escherichia coli as a fusion with maltose binding protein (MBP). The antigenicity of this recombinant MBP-N fusion protein was examined by Western blotting and enzyme linked immunosorbent assay (ELISA). Subsequently, an indirect ELISA was developed to detect rabies specific antibody levels. Using sera from naive and vaccinated animals the ELISA results were compared with virus neutralizing antibodies detected by a rapid fluorescent focus inhibition test (RFFIT). Neutralizing titres by RFFIT were found to correlate well with the OD values in the ELISA (r=0.9436) and the sensitivity and specificity of the ELISA were shown to be 93.4 and 100%, respectively. The data indicate that the recombinant MBP-N fusion protein can be expressed and isolated straightforwardly and may be useful as a safe and abundant source of antigen to monitor seropositivity in vaccinated canines.

Somatotopically arranged inputs from putamen and subthalamic nucleus to primary motor cortex

Employing retrograde transsynaptic transport of rabies virus, we investigated the organization of basal ganglia inputs to hindlimb, proximal and distal forelimb, and orofacial representations of the macaque primary motor cortex (MI). Four days after rabies injections into these MI regions, neuronal labeling occurred in the striatum and the subthalamic nucleus (STN) through the cortico-basal ganglia loop circuits. In the striatum, two distinct sets of the labeling were observed: one in the dorsal putamen, and the other in the ventral striatum (ventromedial putamen and nucleus accumbens). The dorsal striatal labeling was somatotopically arranged and its distribution pattern was in good accordance with that of the corticostriatal inputs, such that the hindlimb, orofacial, or forelimb area was located in the dorsal, ventral, or intermediate zone of the putamen, respectively. The distribution pattern of the ventral striatal labeling was essentially the same in all cases. In the STN, the somatotopic arrangement of labeled neurons was in register with that of corticosubthalamic inputs. The present results suggest that the cortico-basal ganglia motor circuits involving the dorsal putamen and the STN may constitute separate closed loops based on the somatotopy, while the ventral striatum provides common multisynaptic projections to all body-part representations in the MI.

The histopathogenesis of paralytic rabies in six-week-old C57BL/6J mice following inoculation of the CVS-11 strain into the right triceps surae muscle

A fatal encephalomyelitis was developed after intracerebral and hind limb inoculation of in 6-week-old C57BL/6J mice by the inoculation of fixed rabies virus (CVS-11 strain), intracerebrally and into hind. After the intracerebral inoculation, virus antigens were detected in the cerebral cortex and hippocampus at 2 days postinoculation (PI), and later spread centrifugally to thalamus, brain stem, cerebellum, spinal cord and spinal ganglia. At 4 days PI, severe apoptosis and DNA fragmentation were observed in the hippocampus and cerebral cortex. All mice infected intracerebrally were dead without limb paralysis at from 10 to 11 days PI. In contrast, mice infected with virus intramuscularly were persistently observed virus antigens in the myocytes at the site of inoculation from 2 days PI. At 4 days PI, the antigens were demonstrated in the spinal dorsal root ganglia, spinal cord and muscle spindles without their detection in the cerebrum and hippocampus. There were no apoptosis in the spinal cord and dorsal root ganglia, however hind limb paralysis was found in all infected mice. Hind limb paralysis was progressed to quadriparalysis, and mice were dead from 11 to 13 days PI. From 4 days PI, necrosis of neuron was observed in the the spinal and dorsal ganglia with infiltration of lymphocyte. This study suggested that the necrosis of spinal neurons was more important to cause the paralysis of hind limb rather than the severe cerebral infection and apoptosis in C57BL/6J mice infected with CVS-11 strain. The virus primarily replicated in the muscles was ascended the spinal cord via afferent fibers and retrogradely invaded the cerebrum, and with subsequent spread to muscle spindles.

A highly attenuated rabies virus HEP-Flury strain reverts to virulent by single amino acid substitution to arginine at position 333 in glycoprotein

An amino acid at position 333 in the glycoprotein of several fixed rabies virus strains is responsible for the pathogenicity in adult mice. Substitution of arginine at this position largely reduces the viral pathogenicity in adult mice. Attenuation by this single amino acid substitution has been established by using escape mutants selected by monoclonal antibodies and point-mutated virus generated by reverse-genetics. A highly attenuated HEP-Flury strain, which was selected by serial passages in cell cultures, has glutamine at this position. In this study, a point-mutated rHEP333R virus, having arginine at position 333, was generated and examined for the responsibility of this substitution in rabies pathogenicity. The rHEP333R acquired an ability to spread and propagate in mouse brain but the parental rHEP did not. The pathogenicity of rHEP333R to adult mice by intracerebral inoculation largely increased. We confirmed that an arginine at position 333 contributed to reversion of the pathogenicity in a highly attenuated HEP-Flury strain.

Production of rabies neutralizing antibody in hen's eggs using a part of the G protein expressed in Escherichia coli

In an attempt to produce anti-rabies immunoglobulin affordable for people living in developing countries, we have immunized layer chickens with a part of the G protein of rabies virus expressed in Escherichia coli. Immunoglobulin (IgY) was purified from the yolks of eggs layed by immunized hens. It was revealed in vitro that the antibody specifically bound to virions as well as cells infected with rabies virus. Moreover, the antibody apparently neutralized rabies virus infectivity. Inoculation of the antibody into mice infected with rabies virus reduced the mortality caused by the virus, suggesting that IgY directed to the part of the G protein expressed in E. coli could serve as a possible alternative to currently available anti-rabies human or equine immunoglobulins.

Organization of multisynaptic inputs from prefrontal cortex to primary motor cortex as revealed by retrograde transneuronal transport of rabies virus

The organization of multisynaptic projections from the prefrontal cortex to the primary motor cortex (MI) was examined in macaque monkeys by retrograde transneuronal transport of rabies virus. In the first series of experiments, the virus was injected into the MI forelimb region, and the time-dependent distribution patterns of transsynaptic labeling were analyzed in the frontal lobe with various survivals (2-4 d). Two days after the viral injection, neuronal labeling emerged in the caudal aspects of the nonprimary motor-related areas that are known to project to the MI directly. At the same time, the motor thalamus contained labeled neurons. On the third day, cortical labeling extended into the rostral motor-related areas and, also, prearcuate area 8. Moreover, a number of labeled neurons were located in the internal pallidum and the cerebellar nuclei. At the 4 d postinjection period, neuronal labeling occurred widely in prefrontal areas as well as in the putamen and the cerebellar cortex. In the second series of experiments, the viral injection was made into the MI hindlimb region, and the distribution pattern of prefrontal labeling on the fourth day was compared with that in the forelimb-injection case. The labeled neurons in each prefrontal area were much fewer in the hindlimb-injection case than in the forelimb-injection case. Whereas ventral area 46 was most densely labeled from the forelimb region, only sparse labeling from the hindlimb region was observed in this prefrontal area. The present results suggest the importance of ventral area 46 in the cognitive control of forelimb movements.

Generation and characterization of P gene-deficient rabies virus

Rabies virus (RV) deficient in the P gene was generated by reverse genetics from cDNA of HEP-Flury strain lacking the entire P gene. The defective virus was propagated and amplified by rescue of virus, using a cell line that complemented the functions of the deficient gene. The P gene-deficient (def-P) virus replicated its genome and produced progeny viruses in the cell lines that constitutively expressed the P protein, although it grew at a slightly retarded rate compared to the parental strain. In contrast, no progeny virus was produced in the infected host when the def-P virus-infected cells that did not express the P protein. However, we found that the def-P virus had the ability to perform primary transcription (by the virion-associated polymerase) in the infected host without de novo P protein synthesis. The def-P virus was apathogenic in adult and suckling mice, even when inoculated intracranially. Inoculation of def-P virus in mice induced high levels of virus-neutralizing antibody (VNA) and conferred protective immunity against a lethal rabies infection. These results demonstrate the potential utility of gene-deficient virus as a novel live attenuated rabies vaccine.