Fig 1.
Differentiation of neurons after DRG implantation.
Panels A-C, fetal DRG, 18 gestational weeks, prior to subcapsular renal implantation in SCID mouse, stained with (A) hematoxylin & eosin; black arrow denotes nerve root, (B) anti-NCAM (green)/anti-synaptophysin (red) antibody, and (C) subtype specific markers anti-RT97 (green)/anti-peripherin (red) antibody. Panels D-E, DRG xenograft, 20 weeks after transplantation, stained with subtype specific markers anti-RT97 (green)/anti-peripherin (red) antibody (D), demonstrates that neuronal differentiation continues in DRG xengrafts in vivo. Peripherin staining (white arrow) identifies nociceptive neurons; cytoplasmic RT97 immunoreactivity (yellow arrow) identifies mechanoreceptive neurons. Cell counting (E) determined the proportion of nociceptive and RT97-immunoreactive mechanoreceptive neurons at 20 weeks after implantation. For each panel, representative images are shown. For assessment of neuronal subtype at least 20 fields were counted, each with a minimum of 50 neurons.
Fig 2.
DRG architecture is maintained in DRG xenografts.
Panels A-C, DRG xenograft, 20 weeks after transplantation, stained with subtype specific markers anti-RT97 (green)/anti-peripherin (red) antibody (A-C). White arrow in (A) denotes the axon hillock at the neuronal cell body. White dotted line in (C) delineates the margin between the DRG xenograft and the murine kidney, arrow shows axons projecting into the murine kidney. Panel D, transmission electron micrograph of DRG xenograft with arrow showing myelinated nerve fiber. Panel E, DRG xenograft 80 weeks after implantation. Black arrow in (E) denotes the nerve root, panel on the right (F) is inset panel (black box) from (E) with arrows showing small, dark neurons (white) and large, light neurons (yellow).
Fig 3.
VZV-infected DRG xenograft immunostaining at 7 (A), 10 (B-F) and 14 days (G-H) after infection. Representative images are shown for staining experiments, performed on multiple tissue sections (6–10 sections) for each DRG, to assess cytopathology by VZV IE63 protein (A-C, and H), H&E stain (F, G and I), and VZV genomic DNA (D and E). Black arrow in A,B shows IE63 protein in nerve fibers; orange arrow in B,C shows IE63 protein in neuronal cell; yellow arrow in B shows IE63 staining in SGC. Arrow in E,F shows adjacent tissue sections stained for VZV DNA (E) and H&E (F), demonstrating neuronal chromatolysis. Arrow in G shows regions with extensive cytopathic effect; arrow in H shows IE63 negative neuron at 14 days after infection.
Fig 4.
VZV replication is restricted in mechanoreceptive neurons.
Immunostaining VZV-infected DRG xenograft 7–10 days after infection with 330 PFU. Staining experiments were performed on multiple tissue sections (6–10 sections) for each DRG. Panel A, DRG tissue section stained for RT97 marker (green), peripherin (red) and anti-VZV human polyclonal antibody (pink); (B) is inset white panel from A with single channel images to the right. To quantify IE63 positive/RT97 positive neurons compared with IE63 positive/RT97 negative neurons; a minimum of 250 neurons was counted for each condition. Panel C, representative image for cell counting analysis, demonstrating IE63 restriction (green) in RT97 positive (red) neurons (t test p<0.01). (D) Representative image showing extensive IE63 staining (green) in SGC surrounding an RT97-immunoreactive (red) neuron. Single channel images at lower magnification (right) show four numbered neurons in a region with IE63 expression (top) in satellite cells surrounding RT97+ (red) neuron 1 and neuron 3. (E) Cell counting results. (F-G) DRG inoculated with VZV infected T cells (1067 PFU) exhibit the same patterns of spread in xenografts and restricted replication in RT97 immunoreactive neurons. Sections are stained for IE63 (green), and gE (Panel F, red) and RT97 (Panel G, red).
Fig 5.
VZV restriction in mechanoreceptive neurons occurs after viral entry.
For each panel, one representative image is shown for staining experiments performed on multiple tissue sections (6–10 sections) for each DRG. (A-C) ORF23 capsid protein is a marker of virion entry as well as the later formation of progeny virions; virion entry is indicated by ORF23 capsid puncta at the nuclear rim (green), co-stained with cellular lamin A/C (A, red), and prior to expression of IE62 (B, red) or IE63 (C, red). (D) Dual staining for ORF23 (green) and cellular PML (red). (E) ORF23 nuclear rim staining was not observed when using pre-immune serum (green); red is N-CAM staining. (F) ORF23 capsid protein is not detected in uninfected DRG, as shown by staining for N-CAM (red) and absence of ORF23 (green). (G-H) Staining of adjacent tissue sections stained with antibody for ORF23 (green) and IE63 (G, red) or RT97 (H, red). Two neurons are circled and shown in both panels. The single 488-channel images for Panel G, shown in greyscale for better visualization, are provided for the circled neurons (1 and 2). The contrast of the G, neuron 2, ORF23 B&W panel, is enhanced so that the discrete ORF23 particles are easier to observe.
Fig 6.
Restriction of VZV replication in mechanoreceptive neurons is associated with restricted IE62 expression.
As described in the Results, reorganization and disruption of PML nuclear bodies were assessed in relation to expression of VZV ORF61 (A, green), IE62 (E and F, green) and IE63 (G, green); PML staining is in red. Representative images are shown; arrows are described in the text. (D) Temporal expression of IE62 (red) and IE63 (green) protein in the same neuron cell body; (D, right) shows the two neurons in panel (D) at higher magnification. (B,C and H,I) Cell counting analysis of ORF61p (C) and IE62 (I) expression in relation to the RT97 subtype marker. Representative panels are shown for ORF61 protein (B, green) and IE62 (H, green) and RT97 marker (B and H, red).
Fig 7.
Mechanoreceptive neurons are selectively depleted after the acute phase of DRG infection and SGC infection contributes to VZV-induced neuronal cell loss and ganglionic damage.
DRG xenografts were infected with rOka-delta_gI or rOka-gE/deltaCys (1000 PFU), which are deficient for VZV-induced SGC-neuron membrane fusion and spread in DRG, to assess the impact of infection limited primarily to SGC on neuronal survival. Representative images are shown; for each staining condition 6–10 slides were evaluated. (A) rOka-infected DRG at 70 days after infection, stained for subtype markers peripherin (green) and RT97 (red). White arrows indicate Nageotte nodules. (B) rOka-gE/deltaCys infected DRG at 28 days after infection, stained with mouse anti-RT97 (red) and rabbit anti-IE63 (green) and at 56 days after infection stained with mouse anti-RT97 (red) and rabbit anti-peripherin (green). (C) rOka-gE/deltaCys infected DRG at 56 days after infection stained with mouse anti-RT97 (red) and rabbit anti-IE63 (green). (D) rOka-delta_gI infected DRG at 70 days after infection stained with mouse anti-RT97 (red) and rabbit anti-IE63 (green).
Fig 8.
Cellular factors induced in VZV-infected DRG include antiviral and pro-inflammatory cytokines.
Cellular factors made by SGC and other DRG resident cells were profiled in whole tissue lysates. Cytokine concentrations of 19 human proteins were significantly increased (N = 18) or decreased (N = 1) in VZV-infected DRG compared with mock-infected DRG (p <0.05), and 12 are grouped by role: (A) interferon-regulated cytokines, (B) pro-inflammatory cytokines and (C) neuroprotective cytokines. Statistical analyses were performed using GraphPad Prism version 6.0. The complete dataset is shown in S1 and S2 Figs.
Fig 9.
Model for VZV infection of DRG.
Primary VZV infection (A) is initiated in respiratory epithelial cells followed by transfer to T cells in tonsils and other regional lymphoid tissue (B) enabling T-cell mediated spread to skin (C) and subsequently to DRG neurons by retrograde axonal transport (D), or T-cell mediated spread to DRG (E), which facilitates SGC infection. VZV gains access to both nociceptive (small, orange) and mechanoreceptive (large, green) neuron cell bodies by either route; however, replication is severely restricted in mechanoreceptive neurons. If neuronal replication is uncontrolled, infection of SGC facilitates contiguous spread to neighboring NSC (F). Contiguous spread with viral transfer into new neuronal cell bodies amplifies the opportunities for VZV delivery to skin sites of replication. Over time, the consequences of VZV infection in SGC contributes to neuronal cell loss, which is indicated by satellite cell microproliferations, referred to as nodules of Nageotte (G).