Infection with mosquito-borne alphavirus induces selective loss of dopaminergic neurons, neuroinflammation and widespread protein aggregation

Abstract

Neuroinvasive infections with mosquito-borne alphaviruses such as Western equine encephalitis virus (WEEV) can cause post-encephalitic parkinsonism. To understand the mechanisms underlying these neurological effects, we examined the capacity of WEEV to induce progressive neurodegeneration in outbred CD-1 mice following non-lethal encephalitic infection. Animals were experientally infected with recombinant WEEV expressing firefly luciferase or dsRed (RFP) reporters and the extent of viral replication was controlled using passive immunotherapy. WEEV spread along the neuronal axis from the olfactory bulb to the entorhinal cortex, hippocampus and basal midbrain by 4 days post infection (DPI). Infection caused activation of microglia and astrocytes, selective loss of dopaminergic neurons in the substantia nigra pars compacta (SNpc) and neurobehavioral abnormalities. After 8 weeks, surviving mice displayed continued loss of dopamine neurons in the SNpc, lingering glial cell activation and gene expression profiles consistent with a neurodegenerative phenotype. Strikingly, prominent proteinase K-resistant protein aggregates were present in the the entorhinal cortex, hippocampus and basal midbrain that stained positively for phospho-serine129 α-synuclein (SNCA). These results indicate that WEEV may cause lasting neurological deficits through a severe neuroinflammatory response promoting both neuronal injury and protein aggregation in surviving individuals.

Keywords: Neuroimmunology; Parkinson's disease.

Fig. 1

Regional specificity of recombinant Western equine encephalitis virus following intranasal infection. a Schematic diagram illustrating the structure of vectors for each recombinant virus used in these studies. Subgenomic promoter (SPG), untranslated region (UTR). b Schematic diagram illustrating treatment regimen with WEEV. c Raw image of luciferase activity in McFly-infected mice (left-24 HPI, middle-48 HPI, right-72 HPI). d Pseudo-colored image of luciferase activity overlaid onto whole body images during the timecourse of infection. e Ex vivo sagittal hemispheric images of luciferase activity at specific times post infection demonstrating progressive spread of WEEV from the olfactory bulb to caudal brain regions. f Six-week-old CD-1 mice were administered 1 × 10⁴ PFU of DsRed producing WEEV (McRed) via intranasal inoculation and euthanized 4 days post infection. Brains were cryosectioned coronally and dsRed + infected nuclei were labeled (purple) with reference to the allen brain atlas. Individual nuclei with presence of Dsred were labeled: GL (glomerular layer), SI (substantia innominate), MB (mammillary hippocampus (HIP), thalamus (TH), entorhinal regions of the cortex (ENT), and pons

Fig. 2

dsRed-expressing WEEV (McRed) shows strong tropism for dopaminergic neurons and causes selective dopaminergic neuronal loss and glial activation following intranasal inoculation. a–d Neuronal densities in hippocampus and entorhinal cortex were examined at 4 DPI following intranasal infection with McRed. No significant neuronal loss was detected in the hippocampus (CA1) or the entorhinal cortex (338 ± 6.557 vs. 342.8 ± 9.259, p = 0.7142 and 101.3 ± 6.96 vs. 90.4 ± 18.1, p = .6730) (N = 4/group). e–g Sagittal and coronal sections from CLARITY processed brains (h–j) were immunostained for TH + dopaminergic neurons. High-magnification confocal images (k–n) demonstrate co-localization of DsRed (red) and TH + DA neurons (green). White dotted regions delineate the substania nigra pars compacta (SN). n X and Y coordinates in top panel with X and Z coordinates in lower panel. o Number of DA neurons in SNpc were examined with 3D design-based stereology 4 DPI. A 29% loss of DA was observed between treatment groups. (29% loss, 8707 ± 202.2 vs. 6180 ± 162.6). Intense staining for IBA1-positive microglia (green) (p) and GFAP-positive astrocytes (q) was noted at 4 DPI in the SN following infection with McRed when compared with aged matched controls (bottom right of panels). White dotted region delineates the substania nigra pars compacta (SN) (N = 3/group). ****p < 0.0001

Fig. 3

Anti-E1 immunotherapy in WEEV-infected CD-1 mice rescues from lethal infection and facilitates clearance of virus by 8 weeks post infection. a, b Mice were infected with McFly and received either a mock treatment (naive pre-immune serum) or anti-E1 immunotherapy (anti-E1 polyclonal rabbit immune serum). Bioluminescence measurements were acquired every 24 h following inoculation. c Immunotherapy treatment was optimized to achieve consistent CNS infection without mortality, demarcated by red lines indicating the upper and lower range of acceptable luciferase activity. For each treatment regimen, the measured luciferase activity is shown for the head region. Volumes indicate amount of immunized rabbit serum administered. 2X means the animals in that group received doses of immune serum at both 12 and 48 h post infection. d Survival curves for the different treatment regimens employed. e, f Following treatment with immune serum at 12 and 48 h post infection, regimen after intranasal infection with McFly and McRed, animals completely cleared virus by 8 weeks post infection

Fig. 4

Following intranasal infection with WEEV and immunotherapy, surviving mice develop progressive dopaminergic neuronal loss and motor deficits. a Six-week-old CD-1 mice were intranasally (I.N.) inoculated with luciferase-expressing WEEV (McFly) or saline vehicle control and treated with anti-E1 polyclonal antiserum at 12 and 48 h post infection and euthanized 8 weeks later. Outlined regions indicating the VTA and SN were used for quantification. b–g Immunofluorescence of DA neurons (red) and total neurons (green) of control or infected. h, i DA terminals from the striatum of control and infected mice with representative high-magnification images. White dotted line indicates the region used for quantification. j–n Quantitative assessment of DA neurons and total neurons in the SNpc and revealed a 38% loss of TH + dopaminergic neurons (9774 ± 494.5 vs. 6009 ± 374.5) and a 31% loss of NeuN + total neurons (13347 ± 365.6 vs. 9141 ± 436.1) in the SNpc. Dopaminergic neurons in the VTA were decreased by 30% following infection (14,388 ± 1624 vs. 10,002 ± 1026), without significant loss of NeuN + total neurons (33,216 ± 2463 vs. 25,488 ± 2595). n DA terminal mean intensity reduction of 29% when compared with control. o–q Infected mice exhibited an increased run duration when traveling a fixed distance, along with a decreased duty cycle relative to control mice (0.242 ± 0.1912 vs. −0.2618 ± 0.1044 and 0.5509 ± 2.211 vs. 7.383 ± 1.378) when measured with a quantitative gait analysis system. All measurements were compared with baseline control values for each individual mouse (N = 6/group). *p < 0.05, **p < 0.005, ***p < 0.0005, ****p < 0.00005

Fig. 5

Persistent microgliosis and astrogliosis is present following encephalitic infection with WEEV. Six-week-old CD-1 mice were intranasally (I.N.) inoculated with luciferase-producing McFly virus or saline and intraperitoneally treated with anti-E1 polyclonal antibody at 12 and 48 h post infection and euthanized 8 weeks later. a, b White dotted region and gray dotted regions delineate the substania nigra pars compacta (SNpc) and substania nigra reticulate (SNr) used for microglia (IBA1, red) quantification. c, d Iba1-positive microglia were detected by immunofluorescence microscopy in control (A) and infected (B) CD-1 mice 1 week post infection. White dotted region indicates the striatum of control and infected animals. e Quantitative analysis of each respective brain region in control (white) and infected brains (red) revealed a 46% increase of IBA1 + microglia in the SNpc (499.5 ± 52.37 vs. 270.8 ± 42.27), a 31% increase IBA1 + microglia in the SNr (486 ± 42.55 vs. 337.3 ± 45.19), and a 40% increase of microglia in the ST (416.6 ± 36.37 vs. 248.6.3 ± 38.87). f, g White dotted and gray dotted regions delineate the substania nigra pars compacta (SNpc) and substania nigra reticulata (SNr) used for astrocyte (GFAP, green) quantification. c, d White dotted region indicates the striatum of control and infected animals. j Quantitative analysis of infected brains revealed a 69% increase of GFAP + astrocytes in the SNpc (555.1 ± 57.1 vs. 170 ± 40.84), a 56% increase in GFAP + astrocytes in the SNr (882.6 ± 94.91.55 vs. 494.8 ± 102.1), and a 79% increase of astrocytes in the ST (189.8 ± 24.29 vs. 40.07 ± 6.859) (N = 6/group). *p < 0.05, **p < 0.005, ***p < 0.0005, ****p < 0.00005

Fig. 6

Surviving mice show formation of proteinase K-resistant α-synuclein aggregates in multiple brain regions. CD-1 mice were infected with McFly virus and treated with immunotherapy regimen and euthanized 8 weeks post inoculation. a–j Infected and control sections were stained with TH (tyrosine hydroxylase, red), P129 (phosphorylated α-synuclein 129, green), and DAPI 8 weeks post inoculation with McFly. b–e High-magnification images of uninfected control mice and g–j infected brain mice. White dotted region delineates the substania nigra pars compacta (SN). Individual nuclei positive for P129 staining were labeled as follows: HIP (hippocampus), CTX (cortex), and MB (mammillary body). k–n High-magnification images, ×20 magnification, showing P129 + plaques and co-localization with IBA1 + microglia in select nuclei. o, x Proteinase K-treated brain sections with complete degradation of TH depicting proteinase K-resistant α-synuclein aggregates (green). Also depicted are high-magnification images showing the presence of P129 + proteinase K-resistant plaques in infected mice (u–x) and the absence of P129 + plaques in uninfected control mice (p–s)