Fig 1.
ROS is produced in both prion-infected and GDL-exposed COCS.
(A) COCS were exposed to POM1 for 8 h-10 days, and neuronal loss was assessed by NeuN+ pixel coverage. Neuronal loss was significant at 3 dpe. Untreated slices (ut), or slices exposed to pooled IgG, were used as controls (grey). (B) In prion-infected COCS (+) neuronal loss became significant at 45 dpi. Controls (-): exposure to non-infectious brain homogenate. Data were analyzed using a two-tailed t-test; n = 9 biological replicates. (C—D) COCS were exposed to DHE, and the number of positive fluorescent particles was counted per 10x view field. (C) DHE conversion assays were performed at various time points between 1 h and 240 h. Data from three experiments are represented in this graph; n = 27 biological replicates. Slices that were untreated or exposed to pooled IgG were used as controls (grey). ROS production was detectable 4 hours after POM1 exposure and after 1 h with a higher POM1 concentration (335 nM). Here and henceforth data are presented as average ± s.d. and were analyzed by one-way ANOVA with Dunnett’s post-hoc test unless otherwise specified (***: p<0.001; **p<0.01; *p<0.05; n.s., not significant). Results of IgG vs POM1 at 335 nM were analyzed using a two-tailed t-test. (D) COCS were infected with RML (+) or exposed to non-infectious brain homogenate (-). DHE conversion assays were performed at various dpi. ROS production peaked at 38 dpi. Data were analyzed using a two-tailed t-test; n = 27 biological replicates. (E) RML-infected slices (42 dpi) were harvested and analyzed for ROS production by the lucigenin assay in the presence (blue) or absence (black) of DPI. DPI decreased superoxide-induced lucigenin conversion; n = 4 pools of 9 biological replicates. (F) DHE conversion was prominent in the forebrains and cerebella of terminally ill scrapie-infected tga20 mice. Data were analyzed using a two-tailed t-test; n = 4 mice. (G) The prion replication antagonists, PPS, CR and Amph, suppressed ROS production in prion-infected COCS (lower half), but not in GDL-exposed COCS (upper half); n = 27 biological replicates. (H) Treatment with PPS, CR, and Amph did not reduce cell death in GDL-treated COCS; n = 9 biological replicates. Neuronal viability was assessed here and henceforth by measuring the percentage of NeuN+ pixels (ordinate). (I) PPS treatment rescued cell death in RML-infected slices for ≥55dpi (red bars). Data were analyzed using a two-tailed t-test; n = 9 biological replicates.
Fig 2.
ROS scavengers and calpain inhibitors are beneficial to prion-infected COCS and mice.
(A) Lower graph: COCS were inoculated with RML or NBH, cultured for 21 dpi, and then treated with various compounds for 22 days. Upper graph: COCS were cultured for 14 days, incubated with POM1 and treated with the same compounds for 10–14 days. Viability (NeuN) was normalized to COCS exposed to IgG and to non-infectious brain homogenate (upper and lower gray dots, respectively). Calpain inhibitors (Calpeptin), antioxidants (Ascorbate, Isoascorbate, NaC), and superoxide dismutase mimetics (MnTBAP) were neuroprotective in both paradigms (red dots), whereas inhibition of AMPA, kainate receptors (CNQX), NMDA receptors (MK801), a mitochondrial membrane permeability transition pore (methazolamide), caspases (ZVAD), and of ER calcium stores (Dantrolene), (black dots) was ineffective; n = 9 biological replicates. The effects of compounds labeled with “†” on POM1-exposed COCS and zVAD labeled with “§” on RML-infected COCS were reported previously [15], [13] and are reproduced here for convenience. (B) COCS were exposed to NBH (-) or to RML (+) as in (A), and harvested at 39 dpi (n = 3). Prion titers were determined by SCEPA. No reduction in prion infectivity was observed in COCS treated with ascorbate (Asc) or NaC (n = 3 biological replicates). (C) Prion-infected COCS were treated with Asc. Protection was discernible for ≥53 dpi (red bars). Data were analyzed using a two-tailed t-test; n = 9 biological replicates. (D) Survival curves of prion-infected tga20 mice treated with AcHyT (median 95dpi) or vehicle (drinking water; median 88.5dpi). AcHyT treatment increased the incubation time of prion-infected mice (p = 0.0287). Results were analyzed using a Mantel-Cox log-rank test; n = 6 untreated mice and n = 7 AcHyT-treated mice. (E) COCS were treated with antioxidants from 21 dpi onwards (n = 3) and harvested at 40 dpi. Cleaved α-fodrin bands (145 and 150 kDa) were quantified and normalized to GAPDH (right). None of the antioxidants suppressed the prion-induced enhancement of α-fodrin cleavage; n = 3 biological replicates. (F) COCS were cultured for 14 days, treated with scFvPOM1 (400nM) or scFvPOM1 preincubated with recombinant PrP (1200nM) in presence of ascorbate, and harvested at 3 dpe. Cleaved α-fodrin bands (145 and 150 kDa) were quantified and normalized to calnexin (right). Ascorbate did not suppress the GDL induced enhancement of α-fodrin cleavage; n = 3 biological replicates.
Fig 3.
anti-FT antibody POM2 counteracts prion-induced neurotoxicity in COCS.
(A) COCS prepared from tga20 mice were exposed to NBH (gray) or infected with RML prions (white), treated with pooled IgG or POM2 (335nM), and stained for NeuN at 44dpi. POM2 treatment afforded significant neuroprotection against prion infection; n = 9 biological replicates. (B) Prion titers of slice homogenates treated as in (A) were determined by SCEPA. POM2 treatment did not significantly reduce infectivity titers in prion-infected COCS. Data points represent the log of the median infectivity dose per mg of tissue culture (TCID50 mg-1) ± s.d.; n = 4 biological replicates. (C) COCS homogenates were treated as in (A), digested with proteinase K, and probed with the antibody POM1 for PrPSc. While the PrPSc bands between 18–30 kDa were somewhat decreased by POM2 treatment, there was a concomitant increase in higher-molecular PrP-immunoreactive moieties presumably representing SDS-resistant oligomers of PrPSc. Because of the different electrophoretic patterns of POM2-treated samples, we have opted to perform a quantification of the entire lane (right panel). POM2 treatment did not decrease the load of PK-resistant PrPSc; n = 3 biological replicates.
Fig 4.
ER stress is detected in both prion-infected and GDL-exposed COCS.
(A) Tga20 COCS (prion-infected or NBH-exposed) were harvested at 42 dpi, and probed for p-PERK, p-eIF2α, and ATF4 by western blot. Densitometry (normalized to actin) showed a trend towards increased p-PERK (p = 0.14) and significantly elevated p-eIF2α and ATF4 after prion infection. (B) COCS were cultured for 14 days, treated with POM1 or IgG, harvested at 3 dpe, and probed for p-PERK, p-eIF2α and ATF4 by western blot. Densitometry (normalized to actin) revealed increased p-PERK, p-eIF2α, and ATF4. Densitometry data were analyzed using a two-tailed t-test; n = 3 biological replicates.
Fig 5.
The transcriptome of RML-infected and POM1-treated COCS.
(A) The number of genes upregulated or downregulated at various time points after RML infection (left) or exposure to POM1 (right). Upregulation or downregulation was defined as >2-fold change over NBH-exposed (left) or pooled-IgG exposed (right) COCS at the same time point. In both paradigms, the total number of differentially expressed genes increased over time. (B) Overlap of upregulated (left) and downregulated (right) genes in RML-infected vs. POM1-exposed COCS. The two paradigms shared 80% of downregulated and 38% of upregulated genes when comparing 3dpe GDL and 45dpi RML. (C) Correlation coefficients (y-axis) of all genes and genes involved in specific signaling pathways between POM1 exposure at different time points (x-axis) and RML exposure at 45 dpi.