α-Synuclein is a defining, key component of Lewy bodies and Lewy neurites in Parkinson’s disease (PD) and dementia with Lewy bodies (DLB), as well as glial cytoplasmic inclusions in multiple system atrophy (MSA). The distribution and spreading of these pathologies are closely correlated with disease progression. Recent studies have revealed that intracerebral injection of synthetic α-synuclein fibrils or pathological α-synuclein prepared from DLB or MSA brains into wild-type or transgenic animal brains induced prion-like propagation of phosphorylated α-synuclein pathology. The common marmoset is a very small primate that is expected to be a useful model of human diseases. Here, we show that intracerebral injection of synthetic α-synuclein fibrils into adult wild-type marmoset brains (caudate nucleus and/or putamen) resulted in spreading of abundant α-synuclein pathologies, which were positive for various antibodies to α-synuclein, including phospho Ser129-specific antibody, anti-ubiquitin and anti-p62 antibodies, at three months after injection. Remarkably, robust Lewy body-like inclusions were formed in tyrosine hydroxylase (TH)-positive neurons in these marmosets, strongly suggesting the retrograde spreading of abnormal α-synuclein from striatum to substantia nigra. Moreover, a significant decrease in the numbers of TH-positive neurons was observed in the injection-side of the brain, where α-synuclein inclusions were deposited. Furthermore, most of the α-synuclein inclusions were positive for 1-fluoro-2,5-bis (3-carboxy-4-hydroxystyryl) benzene (FSB) and thioflavin-S, which are dyes widely used to visualize the presence of amyloid. Thus, injection of synthetic α-synuclein fibrils into brains of non-transgenic primates induced PD-like α-synuclein pathologies within only 3 months after injection. Finally, we provide evidence indicating that neurons with abnormal α-synuclein inclusions may be cleared by microglial cells. This is the first marmoset model for α-synuclein propagation. It should be helpful in studies to elucidate mechanisms of disease progression and in development and evaluation of disease-modifying drugs for α-synucleinopathies.

The lentiviral vector system is used extensively in gene therapy trials for various neurological and neurodegenerative disorders. The vector system permits efficient and sustained gene expression in many cell types through integration of the transgene into the host cell genome. However, there is a significant issue concerning the therapeutic use of lentiviral vectors, that transgene insertion may lead to tumorigenesis by altering the expression of proto-oncogenes adjacent to the integration sites. One useful approach for improving safety is to restrict vector transduction to neuronal cells. We have reported the use of human immunodeficiency virus type 1 (HIV-1)-based vectors for efficient retrograde transport by pseudotyping with rabies virus glycoprotein (RV-G) or fusion glycoprotein B type, in which the cytoplasmic domain of RV-G was substituted with the counterpart of vesicular stomatitis virus glycoprotein (VSV-G). Here we developed a novel vector system for neuron-specific retrograde gene transfer (termed NeuRet) by pseudotyping the HIV-1 vector with fusion glycoprotein C type (FuG-C), in which a short C-terminal segment of the extracellular domain and the transmembrane/cytoplasmic domains of RV-G were replaced with the corresponding regions of VSV-G. FuG-C pseudotyping caused efficient gene transfer, mainly through retrograde transport, into neuronal cells in diverse brain regions, whereas the pseudotyping resulted in less efficiency for the transduction of glial and neural stem/progenitor cells. Our NeuRet vector system achieves efficient retrograde gene delivery for therapeutic trials and improves their safety by greatly reducing the risk of gene transduction of dividing cells in the brain.

Lines of evidence indicate that both the ventrolateral prefrontal cortex (vlPFC) (areas 45/12) and dorsal premotor cortex (PMd) (rostral F2 in area 6) are crucially involved in conditional visuomotor behavior, in which it is required to determine an action based on an associated visual object. However, virtually no direct projections appear to exist between the vlPFC and PMd. In the present study, to elucidate possible multisynaptic networks linking the vlPFC to the PMd, we performed a series of neuroanatomical tract-tracing experiments in macaque monkeys. First, we identified cortical areas that send projection fibers directly to the PMd by injecting Fast Blue into the PMd. Considerable retrograde labeling occurred in the dorsal prefrontal cortex (dPFC) (areas 46d/9/8B/8Ad), dorsomedial motor cortex (dmMC) (F7 and presupplementary motor area), rostral cingulate motor area, and ventral premotor cortex (F5 and area 44), whereas the vlPFC was virtually devoid of neuronal labeling. Second, we injected the rabies virus, a retrograde transneuronal tracer, into the PMd. At 3 days after the rabies injections, second-order neurons were labeled in the vlPFC (mainly area 45), indicating that the vlPFC disynaptically projects to the PMd. Finally, to determine areas that connect the vlPFC to the PMd indirectly, we carried out an anterograde/retrograde dual-labeling experiment in single monkeys. By examining the distribution of axon terminals labeled from the vlPFC and cell bodies labeled from the PMd, we found overlapping labels in the dPFC and dmMC. These results indicate that the vlPFC outflow is directed toward the PMd in a multisynaptic fashion through the dPFC and/or dmMC.