Fig 1.
Mislocalization of TDP-43 in TMEV-infected cultured cells.
(A, D, E) Immunofluorescent staining of TDP-43, PTB1, and FUS in BHK-21 cells at 8 HPI. (A) TDP-43 is located in the nucleus of mock-infected (VP1-negative) cells. In DA and GDVII infections TDP-43 is depleted from the nucleus and mislocalizes to the cytoplasm of VP1-positive cells where it aggregates (arrows). In contrast, DAΔL and GDVIIΔL infections fail to induce TDP-43 mislocalization. (B) Frequency of TDP-43 mislocalization in TMEV-infected cells at different HPI. DA and GDVII infections induce mislocalization of TDP-43 in almost all VP1-positive cells that begins at least as early as 6 HPI and lasts for at least 12 HPI. In contrast, infection with TMEVΔL virus infrequently leads to TDP-43 mislocalization. (C) pTDP-43 is present in the cytoplasm of DA and GDVII-infected L929 cells (arrowheads). (D, E) PTB1 (D) and FUS (E) are mislocalized to the cytoplasm in DA infection (arrows). (F) 48hs after transfection with pDAL, TDP-43 is mislocalized and aggregates (arrows) in the cytoplasm of L-expressing BHK-21 cells (indicated by Myc positivity). Scale bars: 10 μm. *P < 0.001.
Fig 2.
TMEV infection induces aggresome formation in rodent, but not human cells.
(A) Double immunofluorescent staining for TDP-43 and vimentin in DA-infected BHK-21 cells at 8 HPI. Cells have a large juxtanuclear structure covered by vimentin that represents an aggresome (arrowheads). TDP-43 is localized within the aggresome. (B) DA and GDVII infection in L929 cells leads to cytoplasmic mislocalization and aggregation of PTB1, while DAΔL and GDVIIΔL infections induce minimal mislocalization of PTB1 at this time. (C) HeLa cells infected by DA and GDVII show mislocalization of TDP-43, however, no aggresome is induced (arrowheads). DAΔL and GDVIIΔL infection do not induce TDP-43 mislocalization. Scale bars: 10 μm (A) and 5 μm (B, C).
Fig 3.
The aggresome is important in TMEV replication.
(A, B) DA and GDVII virus-infected BHK-21 cells at 8 HPI. The aggresome in TMEV-infected cells contains VP1, ds-RNA (A) and L (B). L* is expressed in cytoplasm but outside of aggresome (B). (C) ds-RNA is present in small aggregates throughout the cytoplasm of DAΔL-infected BHK-21 cells at 8 HPI. (D) Nocodazole, a microtubule inhibitor, hampers aggresome formation in DA-infected BHK-21 cells at 8 HPI. (E) The amount of DA virus genome is decreased in a dose-dependent manner 6 HPI with two different MOIs in BHK-21 cells that had been treated with nocodazole (0, 3, 10, 30 μM) for 1h prior to infection. (F) Plaque assay of DA-infected BHK-21 cells treated with nocodazole treatment (10 μM, for 1h prior to infection) shows a decrease in virus titer compared to untreated cells. Scale bars: 10 μm.
Fig 4.
Differences in SG formation in TMEV wt and ΔL infection.
(A-C) G3BP1, eIF3A and TIA1 are localized in aggresomes in DA virus-infected BHK-21 (A) and L929, but not HeLa cells (B). (D) G3BP1-positive SGs (arrows) and the aggresome (arrowhead) are rarely found in the cytoplasm of DAΔL-infected BHK-21 cells at 8 HPI. (E, F) Immunofluorescent staining of BHK-21 cells infected with DAΔL and GDVIIΔL virus at 8 HPI. (E) G3BP1-positive SGs are present in the cytoplasm of VP1-positive cells (arrows), while uninfected cells have homogeneous cytoplasmic G3BP1 staining. (F) SGs in GDVIIΔL virus-infected cells contain eIF3A and TIA1 (arrows). (G, H) Following infection with DAΔL and GDVIIΔL, TDP-43 and PTB1 are present in structures that resemble SGs (arrows); however, they infrequently colocalize with SG markers. Scale bars: 10 μm. *P < 0.01, **P < 0.001.
Fig 5.
TMEV infection induces cleavage of TDP-43 and abnormal splicing.
(A) Western blot of BHK-21 cells that are either uninfected or 8 hours after infection with DA, DAΔL, GDVII or GDVIIΔL virus. As a positive control for cleavage of TDP-43, BHK-21 cells were treated with 10 μM MG-132 for 8hrs. Western blots of cell lysates were immunostained with antibody against TDP-43 (C-terminal) and VP1. In addition to the predicted full-length normal 43-kDa band, 35-kDa and 25-kDa bands are prominently seen in the RIPA-insoluble fraction from wt and TMEVΔL virus-infected cells. The levels of full-length and cleaved TDP-43 were quantitated by densitometric analysis using NIH ImageJ and presented under each blot. The value of each of the bands in the MG132-treated or infected cells was compared to the uncleaved band in mock, which was set to 1. (B) Representative agarose gel electrophoresis of RT-PCR products in CFTR minigene-transfected cells that were or were not infected with DA or GDVII virus. Exon 9 included (+) and excluded (–) RT-PCR products are shown. (C) Spliced versus unspliced ratios were calculated and then normalized to the value of L929 cells that received vector, but were not infected. Mean values are from three different experiments performed. *P < 0.001.
Fig 6.
TDP-43 mislocalization and phosphorylation in the CNS 1 week after infection with GDVII virus.
(A, B) TDP-43 is depleted from the nucleus of motor neurons in the anterior horn of an ALS patient, and is phosphorylated and mislocalized into the cytoplasm into skein-like inclusion. (C-O) GDVII virus-inoculated mice at 1 week following infection. (C) immunohistochemical staining for VP1 shows GDVII-infected hippocampal CA1 neurons and axons (inset). TDP-43 is mislocalized within the cytoplasm of infected hippocampal CA1 neurons (D, E), but not in uninfected CA3 neurons (F). pTDP-43 is present in the nucleus (G) and cytoplasm (H) of infected hippocampal neurons. (I) pTDP-43 is present as a small aggregate in infected brainstem neurons. (J) Haematoxylin and eosin (HE) staining shows perivascular inflammatory infiltrates in the lumbar spinal cord. (K, L) Numerous VP1-positive cells are observed in the anterior horn at two magnifications. (L-O) Higher magnification of the anterior horn shown in I. (L) Immunoreactivity for VP1 is seen in motor neurons and axons. (M) Expression of TDP-43 in the nucleus of anterior horn cells is decreased and mislocalized to the cytoplasm. (N) PTB2 is depleted in the nucleus of anterior horn cells (arrow). (O) TDP-43 is present in the nucleus of an uninfected cell, while a VP1-positive motor neuron has decreased TDP-43 staining in the nucleus with cytoplasmic round and linear aggregates (arrows). Scale bars: 200 μm (C, D, J, K), 50 μm (E, F), 10 μm (G-I, O) and 20 μm (A, B, L-N).
Fig 7.
TDP-43 mislocalization and phosphorylation in DA virus-infected mice.
(A-E) Serial sections of hippocampus two weeks following infection with DA virus (acute phase). (A-B) Neurons of the CA2 hippocampal region are infected, as indicated by VP1 positivity. (C, D) The expression of TDP-43 and PTB2 is decreased in the nuclei and cells in the infected region. (E) pTDP-43 is present in the nucleus and cytoplasm of infected CA2 neurons. (F-K) 6 weeks following infection with DA virus (chronic phase). (F-H) Serial sections of thoracic spinal cord. (F) HE stain shows perivascular infiltrates and vacuolation in the ventral part of the cord (arrowheads). (G) Immunostaining for CNPase shows demyelination. (H) Immunostaining for Iba-1 shows accumulation of activated microglia and macrophages in the demyelinated region. (I) In the demyelinated region, immunoreactivity for VP1 is present in glial cells and myelin, including the cytoplasm of an oligodendrocyte (inset). (J) TDP-43 is mislocalized to the cytoplasm in glial cells in the demyelinated region (arrowheads). (K) Immunofluorescent staining for CNPase, VP1 and TDP-43 in the demyelinated region. A VP1-positive oligodendrocyte shows depletion of TDP-43 from the nucleus and mislocalization to the cytoplasm. Note the normal TDP-43 nuclear staining pattern in uninfected cells (arrowheads). Scale bars: 50 μm (A-E), 200 μm (F-H), 20 μm (I, J) and 5 μm (K).