JCS043604 Supplementary Material

Files in this Data Supplement:

• Supplemental Figure S1 -

  Fig. S1. Serum deprivation caused increased intracellular ROS but not cell death in CF10 and N2a cells. The stress induced by serum withdrawal was assessed for CF10 (A,B) and N2a (C,D) cells by progressively depleting media serum. The DCFDA assay for intracellular reactive oxygen species was used to monitor cellular stress response (A,C). Serum deprivation induced a progressive increase in intracellular reactive oxygen species to a maximum response seen when no serum was present. This response was significant by one-way ANOVA for both CF10 and N2a cell lines; F=22.33, P<0.0001, n=4 and F=5.365, P=0.0044, n=4 respectively, * P<0.05, *** P<0.001. Cell viability, measured by the MTS assay, showed that despite the increase in intracellular ROS the cells responded to the serum deprivation by increasing their growth rate as opposed to showing reduced viability (B,D). The increased viability in response to serum withdrawal is not significant for the CF10 cells (one-way ANOVA; F=2.040, P=0.0839, n=4) but reaches significance in the N2a cells (one-way ANOVA; F=2.855, P=0.0220, n=4; *P<0.05).

• Supplemental Figure S2 -

  Fig. S2. A scrambled PrP23-89 peptide does not alter intracellular ROS induced by serum deprivation in CF10 and N2a cells. To determine the specificity of the PrP23-89 modulation of intracellular ROS, the DCFDA assays performed on the CF10 and N2a cells with the PrP23-89 peptide were repeated using a peptide consisting of the amino acid composition of PrP23-89 scrambled (PrP23-89scram). (A) ApoPrP23-89scram showed no ability to alter the levels of intracellular ROS induced by serum deprivation at any of the concentrations tested (F=0.8359, P=0.5736, n=4). (B) Applying the peptide with equimolar then 2-6 molar equivalents CuCl-6×glycine did not reproduce the ROS attenuating effect of the copper-loaded PrP23-89 peptide. The profile obtained is significantly different from the PrP23-89 results using two-way ANOVA but not significantly different from the copper-alone profile (F=21.99, P<0.0001, n=4 and F=0.01927, P=0.8903, n=4 respectively). (C) When the apoPrP23-89scram was applied to the N2a cells with and without full copper loading no significant change from the baseline serum depleted ROS response was observed (one-way ANOVA; F=1.691, P=0.2616, n=4).

• Supplemental Figure S3 -

  Fig. S3. Filipin III complex treatment does not alter location or internalisation of the lipid-raft-excluded transferrin receptor. To assess whether the Filipin III complex treatment, which disrupts lipid-raft domains by sequestering cholesterol, might be altering non-lipid raft functions at the membrane, cells were assessed for the membrane versus internal localisation of the transferrin receptor. The transferrin receptor is excluded from lipid-raft domains and so should give an indication of any confounding effects of the filipin III treatment on non-lipid-raft domains within the cell membrane. Cells were incubated in Opti-MEM I media with or without the addition of 1 µM filipin III complex for 30 minutes, a time point chosen as the effects of the peptides are beginning to become apparent in the rate curves (Figure 1B) and any accumulation of the receptor at the cell surface or internally should be apparent. After the incubation period half of the cells were treated with trypsin to remove cell surface protein and the remaining half were incubated in dPBS. Cells were then harvested by centrifugation and lysed in 50 µl RIPA buffer. Lysates were western blotted as described previously using 1 in 1000 dilution of anti-transferrin receptor antibody (Invitrogen), and 1 in 5000 secondary anti-mouse HRP antibody. At this high concentration of secondary the murine IgG heavy chain is observed in the blots running at a slower mobility than the transferrin receptor. Following stripping in 1% (v/v) HCl, β-tubulin (Sigma-Aldrich; at 1 in 10,000, with 1 in 10,000 anti-mouse HRP secondary antibody) was blotted as a control for the concentration of total/internal protein detected. (A) Representative western blots of cell lysates. (B) Densitometric quantification shows that filipin III treatment does not alter overall levels of the transferrin receptor (Student’s t-test, t=0.8284, P=0.4945, n=3) and (C) the receptor is not seen to have altered internalisation as a result of filipin III treatment compared to the untreated cells (Student’s t-test, t=0.4721, P=0.669, n=3).

• Supplemental Figure S4 -

  Fig. S4. Heparin lyase III digestion removes greater than 40% of cellular heparin sulphate. Cells were incubated in Opti-MEM I media for 1 hour with or without the addition of 10 mU/ml heparin lyase III. After the incubation period cells and media were separated by centrifugation, and cells were lysed in 50 µl RIPA buffer. To determine the efficiency of digestion a dot blot for total heparin sulphate was prepared for both the cell lysate and media. One hundred micrograms of cell lysate protein or 5 µl of media was spotted onto a nitrocellulose membrane, dried at 37°C for 1 hour, then blocked and blotted using 10E4 primary antibody (Seikagaku; 1 in 1000 dilution) and anti-mouse HRP-conjugated secondary antibody (Amersham; 1 in 1000 dilution). The 10E4 antibody detects partially, but not fully, degraded heparin sulphate and therefore can detect heparin sulphate in the cell lysate and media. (A) Representative dot blots of cell lysates and media; lower signal intensity is observed after digestion. (B) Densitometric quantification indicates that over 40% of cellular heparin sulphate has been degraded by the heparin lyase III. This is significant for both the cell lysates and the media used for digestion (F=7.285, P=0.0003, n=4; *P<0.05, **P <0.01).

• Supplemental Figure S5 -

  Fig. S5. CD spectra of PrP23-50 P26/28A. Both PrP23-50 (dotted line) and PrP23-50 P26/28A (solid line) adopt a predominantly random coil structure. The apparent magnitude of the mean residue elipticity for PrP 23-50 P26/28A is reduced due to its increased tendency to aggregate.

• Supplemental Figure S6 -

  Fig. S6. The effect of copper saturation on the cellular association and degradation of PrP23-89 P26/28A. 10 µM of the PrP23-89 P26/28A fragment, with and without premixing with 4 molar equivalents CuCl₂-6×glycine, and the PrP23-50 fragment were added to cells at 0 (background control) to 60 minutes. After this time cell lysates as well as media were harvested and western blotted using mab 8B4 (against amino acid residues 37-44) to detect the fragments. Example blots are shown in Figure 7D. (A,B) The cell-associated fraction, (C,D) the media fraction and (E,F) the sum of both fractions as an indicator of total loss of the peptide from the system, of the apo-23-89 P26/28A (A,C,E) and the copper-loaded PrP23-89 P26/28A (B,D,F). Data points are the mean and s.e.m. derived from three independent experiments. Where appropriate, the single exponential rate curves for association or decay are shown (unbroken line) with the 95% CI (broken lines). As the PrP23-89 P26/28A peptide aggregates, with two dominant oligomeric species appearing above the monomer, each of these three bands has been quantified separately and the upper, middle and monomer (lower) band data is shown (on the same scale as the total intensity data) to the right of the corresponding graph. Copper binding reduces the rate of cellular association of the PrP23-89 P26/28A fragment and the intensity once association has reached equilibrium, and, in contrast to the PrP23-89 fragment, increases its loss from the media and the overall system. The monomer behaves differently to the oligomeric bands both as regards cellular association and reduction/loss from the media and overall system.