Fig 2: Changes in levels of VAPB and PTPIP51 in temporal cortex correlate significantly but loss of VAPB does not correlate with ER PDI levels and loss of PTPIP51 does not correlate with mitochondrial MCU levels in Alzheimer's disease. NSE-normalised VAPB, PTPIP51, PDI and MCU signals from immunoblots were correlated by generating correlation coefficients and significance was established by non-parametric, two-tailed Spearman Rho tests. Graphs show normalised levels of VAPB and PTPIP51 (A), VAPB and PDI (B) and PTPIP51 and MCU (C) in the different samples. Each point in the graphs represents data from an individual case. Overall, VAPB and PTPIP51 levels displayed significant correlation (r = 0.57,p < .0001). However, whilst the control group showed positive correlation (r = 0.68, p = .009) the correlation was lost in the Alzheimer's groups (Braak stage III-IV, r = 0.16, p = .59; Braak stage VI r = 0.42, p = .14). A positive correlation between VAPB and PDI was detected in control cases (r = 0.61, p = .02) but this was lost in Braak stage III-IV (r = -0.24, p = .41) and Braak stage VI cases (r = 0.05, p = .88) cases. There was no correlation between PTPIP51 and MCU in control (r = 0.52, p = .06), Braak stage III-IV (r = 0.29, p = .32), or Braak stage VI cases (r = 0.3, p = .3) (Fig. 3C).

Fig 3: Inhibition of protein disulfide isomerase reverses the enhancing effect of Eap on staphylococcal internalization. Internalization of S. aureus strain 8325-4 and S. carnosus TM300 as indicated in EA.hy926 cells in absence and presence of 20 µg/mL recombinant Eap. Cells were pre-incubated with (A and B) bacitracin (10 mM), (C and D) dithiobis-nitrobenzoic acid (DTNB, 10 mM), (E) DMSO (1% vol/vol), (F) anti-PDI antibody RL90 (10 µg/mL), or (G) anti-PDI antibody RL77 (1:100) 30 min before stimulation with Eap or (H and I) PDI expression silencing by P4HB siRNA knock down. In the case of bacitracin, host cells were washed before the addition of bacteria to remove bacitracin. Cells were incubated with an MOI of 50. One hour postinfection, extracellular staphylococci were removed by lysostaphin treatment. Intracellular bacteria numbers were assessed by plate counting. Bacterial numbers in control cells were set to 100%. Recombinant Eap data represent the means ± SD of at least three independent experiments. ***, P = 0.001; **, P = 0.01; *, P < 0.05; one-way ANOVA followed by Bonferroni post-test. (J) siRNA P4HB knockdown (10 pM siRNA) was confirmed by Western blotting. Data represent means ± SD from Western blot quantification of three independent experiments. ***, P = 0.001; one-way ANOVA followed by Bonferroni post-test.

Fig 4: Eap promotes protein disulfide activity on the surface of eukaryotic cells. (A) Eap-stimulated abundance of free ecto-sulfhydryls on the surface of HMEC-1 and (B) EA.hy926 cells was detected with the thiol-reactive dye Alexa Flour 488 C5 Maleimide and measured by flow cytometry. Recombinant Eap data represent the means ± SD of three independent experiments. ***, P ≤ 0.001; *, P < 0.05; one-way ANOVA followed by Dunnett’s multiple-comparison test. (C) Inhibition of Eap-promoted abundance of ecto-sulfhydryls on HMEC-1 by P4HB siRNA (small interfering RNA) knockdown (20 pM siRNA) or anti-PDI antibody RL90 (10 μg/mL) as detected with Alexa Flour 488 C5 Maleimide. Native Eap data represent the means ± SD of three independent experiments. **, P ≤ 0.01; ***, P ≤ 0.001; two-way ANOVA followed by Bonferroni post-test. (D) Bacterial supernatant stimulated (protein disulfide isomerase) PDI activity on the surface of HMEC-1 cells as detected with Alexa Flour 488 C5 Maleimide. Data represent the means ± SD of three independent experiments. **, P ≤ 0.01; two-way ANOVA followed by Bonferroni post-test. (E) Influence of Eap on kinetics of di-E-GSSG (di-eosin-oxidized-glutathione) reduction to E-GSH catalyzed by soluble PDI (200 nM) or (F) in situ-activity in HMEC-1 determined by measurement of fluorescence. Recombinant Eap data represent the means ± SD of three independent experiments. ***, P ≤ 0.001; **, P ≤ 0.01; *, P < 0.05; two-way ANOVA followed by Bonferroni post-test, compared to 0 μg/mL Eap. (G) Impact of bacitracin (10 mM), 5,5′-dithiobis-(2-nitrobenzoic acid) (DTNB, 10 mM), anti-PDI antibody RL90 (10 μg/mL), or P4HB siRNA knockdown (20 pM siRNA) on Eap (20 μg/mL natives Eap) promoted di-E-GSSG reduction to E-GSH in HMEC-1 cells, determined by measurement of fluorescence. Data represent the means ± SD of four independent experiments. ***, P ≤ 0.001; **, P ≤ 0.01; *, P < 0.05; one-way ANOVA followed by Dunnett’s multiple-comparison test (compared to 20-μg/mL control).

Fig 5: Hypothetical model of how Eap mediates the internalization of S. aureus and other bacteria into host cells. Integrin a5ß1 on the host cell is not in an active state. Eap stimulates PDI, which leads to increased activation of a5ß1 integrins. This enhances the binding of Fn. S. aureus binds to Fn via Fn-binding proteins (FnBPs). Eap binds to both S. aureus and Fn and therefore, in concert with FnBPs, may facilitate the binding of S. aureus to a5ß1-integrin via the Fn bridge. Enhanced binding of Fn leads to increased integrin clustering, which results in elevated uptake of S. aureus into the host cell. In the case of S. carnosus and also other bacteria that are normally only poorly taken up by host cells, the increased integrin clustering caused by Eap presumably induces a more vivid internalization activity of ECs, accompanied by Eap bridges between the bacterium and Fn bound to a5ß1-integrin that enables the bacterial uptake.