Mutant-VSV can effectively infect neurons in vivo
Based on the characteristics of rapid amplification, high infection efficiency and expression level of exogenous genes, VSV has been developed into an important gene transfer vector [7, 19]. It has been previously reported that rearrangement or mutation of viral genes in the genome can reduce the replication speed or toxicity [30-34]. Here, recombinant VSV-NR7A was constructed (Fig. 1). The wild-type and mutant VSV with EGFP reporter were recovered and amplified in BHK21 cells (Fig. 2A). WT-VSV and Mutant-VSV were added to BHK-21 cells at an MOI (multiplicity of infection) of 0.001. When infected with WT-VSV, obvious fluorescence signals can be detected within 12 hours, while Mutant-VSV need 24 hours (Fig. 2B). Significant cytopathic changes were observed after 48 hours in WT-VSV group, however, cytopathies were observed 72 hours later in VSV-NR7A group (Fig. 2B). Compared to WT-VSV, Mutant-VSV exhibit a significant reduction (three orders of magnitude) in virus replication at 35 ℃ within 24 hpi (Fig. 2C), which is consistent with previous findings [24]. In addition, the titer of Mutant-VSV increased with time and reached its peak in 72 hours, which was one order of magnitude less than that of wild-type VSV (Fig. 2C). The ability of the Mutant-VSV to infect neurons in vivo was investigated. The Mutant-VSV was injected into the DG region of mouse brain. The brain slices were prepared at 1 DPI and stained with Cy3-conjugated NeuN antibody. Results showed that EGFP fluorescent signals from Mutant-VSV were co-localized with neurons in vivo (Fig. 2D).
Neuroinflammatory responses induced by the WT-VSV and Mutant-VSV vectors
Viral infections can cause neuroinflammation and induce microglia activation [35, 36]. To evaluate microglial activation induced by WT-VSV and Mutant-VSV infection at the injection site, immunohistochemistry for Iba1, the microglial marker, was performed. As shown in Figure 3, microglial infiltration was markedly increased at the injection site (DG region) following WT-VSV injection, while microglial activation due to Mutant-VSV infection was significantly milder even after 3 days (Fig. 3), indicating that Mutant-VSV is less toxic in the injection site.
Anterograde trans-synaptic labeling with Mutant-VSV vector
VSV has the ability of anterograde trans-synaptic tracing of neural circuits (Fig. 4A), and is used for the output network analysis of specific brain region [37]. To examine anterograde trans-synaptic ability of Mutant-VSV, the Mutant-VSV and WT-VSV were injected into the VTA region of mouse brain, and the brain slices were prepared at 3 DPI and 5 DPI. Green signals were observed in injection site (Fig. 4B). Several brain regions including anterior olfactory nucleus (AON), lateral septal nucleus (LS), nucleus accumbens (ACB), bed nuclei of the stria terminalis (BST), habenula, and dorsal nucleus raphe (DR) were labeled by Mutant-VSV at 3 DPI. However, signals were detected in additional brain regions at 5 DPI, including medial preoptic nucleus (MPN), posterior hypothalamic nucleus (PH), dorsomedial nucleus of the hypothalamus (DMH), periaqueductal gray (PAG), superior central nucleus raphe (CS) (Fig. 4C). These results showed that Mutant-VSV has anterograde trans-synaptic ability, and the trans-synaptic efficiency increases with time. For these brain regions, except MPN, the others are the direct downstream regions of VTA projection according to Allen Mouse Brain Connectivity Atlas (http://connectivity.brain-map.org/). These results suggested that Mutant-VSV may anterogradely label the postsynaptic neurons through one synaptic connection at 3 DPI.
Moreover, more brain regions were labeled by wild-type VSV at 5 DPI, including taenia tecta, dorsal part (TTd), diagonal band nucleus (NDB), olfactory tubercle (OT), caudoputamen (CP), somatosensory areas (SS), globus pallidus, external segment (GPe), substantia innominata (SI), fundus of striatum (FS), magnocellular nucleus (MA), anterior amygdalar area (AAA), anterior hypothalamic nucleus (AHN), central amygdalar nucleus (CEA), basomedial amygdalar nucleus, anterior part (BMAa), retrosplenial area (RSP), parafascicular nucleus (PF), ammon's horn Field CA3 (CA3), temporal association areas (TEa), and piriform area (PIR). These results showed that Mutant-VSV had delayed anterograde trans-synaptic ability (Fig. 4C).
Attenuated lethality of Mutant-VSV in mice brain compared with WT-VSV
Rapid lethality in experimental animals is a limitation for most neurotrophic viruses in neuroscience applications [38]. It is important to determine the survival time of mice infected with neurotrophic virus. Eight-week-old C57BL/6 mice were divided into three groups with 10 mice in each. WT-VSV, Mutant-VSV and PBS were injected intracranially into the DG of hippocampus. WT-VSV was rapidly lethal within 1 week and more than half of the deaths occurred at 4 DPI, which was delayed to 14 DPI in Mutant-VSV group (Fig. 5). The survival percentage was analyzed by Log-rank test (P < 0.0001, Fig. 5). These data showed that as compared with WT-VSV, the time of death in Mutant-VSV infected mice was delayed significantly.
More connected brain network were labeled with Mutant-VSV
As the rapid death of experimental animals induced by neurotropic virus, the connected brain networks of injection site was not fully resolved. To determine whether the attenuated virus could label more connected network with the extension of survival time, Mutant-VSV were injected into the VTA of C57BL/6 mice, and the brain slices were imaged at 10 days post-injection. Normal cell morphology of labeled VTA output neurons can be observed at 10 DPI by using Mutant-VSV (Fig. 6A). Moreover, more connected downstream brain regions, which were not labeled at 5 DPI by WT-VSV (Fig. 4C), can be revealed at 10 DPI through Mutant-VSV (Fig. 6B). These results suggested that Mutant-VSV could be used to reveal longer range of downstream networks compared with WT-VSV.