Fig 1.
No change in mRNA expression for TDP-43 species due to LPS treatment.
(A) Representative images of primary astrocytes and microglia cultures double stained for Iba 1(in red, marker of microglia) and GFAP (in green, marker of astrocytes). Nuclei of all cells were stained with DAPI (in blue). Using this double immunostaining we confirmed that we achieved a good separation of astrocytes from microglia, and resulting cultures were ~ 90% pure in one of the cell types. (B) Total RNA was extracted from LPS-treated (1ug/ml LPS for 1 day) and untreated cultures of astrocytes and microglia. Samples of total RNA were then subjected to real-time quantitative RT-PCR for human TDP-43 (hTDP-43, transgene) and mouse TDP-43 (mTDP-43, endogenous gene). Number of copies of hTDP-43 and mTDP-43 were normalized with the house-keeping gene Atp5 mRNA, in order to take in account variations in samples size. Levels of RNA in the LPS-treated sample were divided by the levels of mRNA in the respective untreated control, in order to get the fold changes in the expression with the LPS-treatment. Fold change equal to 1 indicate no effect of LPS in the expression of the gene, which was seen in most of the cultures analyzed for both mouse and human TDP-43.
Fig 2.
Higher levels of TDP-43 protein in LPS-activated astrocytes and microglia.
Cultured astrocytes and microglia from transgenic (hTDP-43A315T) and non-transgenic litters were treated for 1 day with LPS at 1 μg/ml or 0.1 μg/ml. Total protein was extracted from LPS -treated and untreated cultures and analysed by immunoblotting. TDP-43 was detected using the polyclonal antibody (Proteintech # 10782-AP) which reacts with both human (transgenic) and mouse (endogenous) proteins. TDP-43 bands were normalized against actin to take into account the difference in protein loading. The exposure times are actually different between transgenic and non-transgenic blots. Due to higher quantities of TDP-43 (human plus mouse) in the transgenic cell cultures the exposure time was only 30 seconds whereas for the non-transgenic cultures it was 4 minutes. Fold change is the ratio of band intensity in LPS-treated cultures over the band intensity in the respective untreated control. (A) Representative western blot of TDP-43 from LPS treated or untreated astrocyte culture. (B) Quantitative analysis of western blot showed that levels of total TDP-43 was higher in LPS treated transgenic astrocyte culture than non-transgenic culture. (C) Representative western blot of TDP-43 from LPS treated or untreated microglia culture. (D) Quantitative analysis of western blot showed that levels of total TDP-43 were also higher in LPS-treated microglia from transgenic TDP-43A315T mice than from C57BL6 mice.
Fig 3.
Cytoplasmic increase of TDP-43 in LPS-activated astrocytes.
Representative images of astrocytes from hTDP-43A315T transgenic and non-transgenic littermates, treated or not treated with LPS (1 μg/ml) for one day. Astrocytes are double stained for TDP-43 (red) and GFAP (green) while nuclei are stained with DAPI (blue). Immunofluorescence for TDP-43 increased in the cytoplasm of LPS-treated astrocytes in both non transgenic (2nd panel) and transgenic culture (4th panel) as compared to the untreated control (1st and 3rd panel). However, the overall cytoplasmic increase of TDP-43 was more pronounced in LPS-treated astrocytes from hTDP-43A315T transgenic mice than in LPS-treated astrocytes from C57Bl6 mice. Arrows point on cells presenting a decrease in nuclear TDP-43. Scale bar 100μm.
Fig 4.
Cytoplasmic aggregates of TDP-43 in LPS-activated microglia.
Representative images of microglia from hTDP-43A315T transgenic and control littermates, treated or not treated with LPS (1μg/ml) for one day. Microglial cultures were double stained for polyclonal TDP-43 (red) and CD11b (green) whereas the nuclei were stained with DAPI (blue). No cytoplasmic aggregates of TDP-43 were found in untreated microglia from control C57Bl6 mice (1st panel) or from hTDP-43A315T transgenic mice (3rd panel). Treatment with LPS resulted in formation of small cytoplasmic TDP-43 aggregates in microglia from control C57Bl6 mice (2nd panel) and from hTDP-43A315T transgenic mice (4th panel). However, TDP-43 punctate aggregates were more abundant and of higher intense staining in LPS-treated microglia (4th panel) from hTDP-43A315T transgenic when compared to microglia from C57Bl6 mice (2nd panel). Arrowheads point to cells with cytoplasmic aggregates of TDP-43.
Fig 5.
Translocation of TDP-43 from the nucleus to the cytoplasm in astrocytes and microglia with LPS-treatment.
(A) Nuclear and cytoplasmic protein fractions of astrocytes were analyzed by Western blot. The intensity of the 43kDa TDP-43 bands was divided by the intensity of the respective actin band to take in account the differences in the protein loading. Fold change is the ratio of the 43 kDa band intensity in the LPS-treated cultures over the band intensity in the respective untreated controls, after normalized by the intensity of the correspondent actin bands.
Fig 6.
LPS treatment caused changes in nuclear and cytosolic TDP-43 immunostaining intensities in microglia cultures.
The staining intensities were measured using the software Imaris on images taken with the confocal microscope Olympus IX81 and the software Olympus Fluoview ver 3.1a. More than 50 cells were analysed per cultures, 3 cultures of transgenic microglia and 3 cultures of non-transgenic microglia (N = 3). Fold changes were calculated by dividing the average intensities in the LPS-treated cultures over the average intensities in the untreated cultures. Graphic bars represent average and standard error of the mean values calculated with software GraphPad Prism 5.
Fig 7.
TNF-α increases cytoplasmic localization of TDP-43 in NSC-34 cells.
Non-transfected NSC-34 cells and hTDP-43WT-HA stably-transfected NSC-34 cells were treated with recombinant TNF-α at a concentration of 10 ng/mL for 6h. Non-TNF treated cells served as controls. Following treatment, the subcellular distribution of TDP-43 was determined by Western blot analysis. Immunoblotting with polyclonal TDP-43 antibody showed increased cytoplasmic TDP-43 in TNF-α-treated NSC-34 cells (left first panel) and hTDP-43WT-HA stably transfected cells (right first panel). Similarly, immunoblotting with monoclonal TDP-43 antibody (specific for human TDP-43) showed increased levels of cytoplasmic TDP-43 in TNF-α-treated hTDP-43WT-HA stably-transfected NSC-34 cells (2nd left panel). Immunoblots were reprobed with P84 antibody (nuclear marker) and GAPDH antibody (cytoplasmic marker).
Fig 8.
Chronic LPS treatment exacerbated cytoplasmic mislocalization and aggregation of TDP-43 in hTDP-43A315T transgenic mice.
Non-transgenic and transgenic mice expressing hTDP-43A315T mutant at 6 months of age were i.p. injected with LPS (1mg/kg) or vehicle solution once per week up to 2 months. (A) Representative images of spinal cord sections of non-transgenic and hTDP-43A315T transgenic mice, vehicle and LPS treated were double stained with polyclonal TDP-43 (green) and Neu N (red). (B) Measurement of cytoplasmic to nuclear ratio of TDP-43 staining showed increased ratio in LPS treated transgenic as well as non-transgenic mice (N = 4 per group, 8 sections of spinal cord from 4 different animals for each group, *<0.05, **<0.001 by student’s t test.) (C) Chronic treatment of hTDP-43A315T transgenic mice with LPS enhanced levels of insoluble TDP-43 in spinal cord. Total protein was extracted from spinal cords of LPS or vehicle-treated mice and sub-fractionated into insoluble and soluble fractions. Sub-fractionated samples were then analyzed by Western blot using the polyclonal TDP-43 antibody. (D) The intensity of TDP-43 bands were divided by the intensity of the respective actin band to take in account the differences in the protein loading and then the fold change was calculated. The fold change is the ratio of the band intensity in the LPS-treated mice over the band intensity in the respective vehicle-treated controls. Fold changes are all higher than 1, indicating that LPS treatment led to increased levels of TDP-43 protein. Groups were compared using t-test. *p value = 0.03 by student’s t test (N = 4 per group, spinal cord from 4 different animals were used to extract protein for each group). Scale bar 50μm.
Fig 9.
Chronic LPS treatment enhanced nuclear exclusion of TDP-43 in spinal neurons of hTDP-43A315T transgenic mice.
Transgenic mice expressing hTDP-43A315T mutant at 6 months of age were i.p. injected with LPS (1mg/kg) or vehicle solution once per week for a period of 2 months. (A) Spinal cord sections from LPS and vehicle-treated mice were immunostained with monoclonal antibody specific for human TDP-43. As revealed by immunostaining, more neuronal cells in LPS-treated spinal cord sections exhibited loss of nuclear TDP-43 staining (first and second panels) than vehicle (third and fourth panels). Insets in the first and second panels show magnified motor neurons of LPS-treated mice with nuclear loss of TDP-43. Arrow heads in first and second panels show neurons exhibiting nuclear loss of TDP-43 in LPS-treated hTDP-43A315T tg mice while arrows in third and fourth panel show neurons exhibiting nuclear loss of TDP-43 in vehicle-treated hTDP-43A315T tg mice. All representative sections are from different mice for each treatment group. DAPI was used to stain nuclei. Scale bar = 50μm. (B) Number of spinal neurons per section exhibiting nuclear loss of TDP-43 was higher in LPS-treated mice as compared to vehicle-treated mice (p = 0.043 by student’s t test, N = 4, 6 spinal cord sections from 4 different animals at 8 months were analyzed for each group).