Fig 2: Apo-NOX5 persists under heme-depleted conditions and is able to bind exogenous heme. A) Cellular heme conditions probed using a genetically encoded heme sensor HS1 [57] and flow cytometry. The HS1 sensor contains a heme binding moiety, an eGFP sensitive to heme binding, and an mKATE2 that is insensitive to heme binding. The ratio of eGFP:mKATE2 fluorescence is indicative of labile heme content of the cell, with the ratio inversely correlated with heme concentration. The flow cytometry data are representative of HEK293 cells cultured in normal media (black), heme-depleted media treated with succinyl acetone (SA) to inhibit heme synthesis (gray) and heme-depleted media with SA and 5 µM hemin added for 2 h (red). B) Representative western blot analysis of HEK293 cells and HEK293 cells stably expressing NOX5ß grown in either normal media or heme-depleted media with SA and quantification of expression in these conditions. Blots correspond to panels C and D. Quantification values represent means ± SEM, n = 6, *** denotes p < 0.001, two-tailed t-test. C-E) Reduced minus oxidized difference spectra of live cells using the Olis Clarity. After addition of cycloheximide to halt protein synthesis, cells were normalized to live cell number and the UV–vis spectrum was taken before and after addition of sodium dithionite. The oxidized spectrum was then subtracted from the reduced spectrum. C) Difference spectra of HEK293 cells (green) and HEK293 cells stably expressing NOX5ß (blue). Characteristic NOX5 peaks are visible at 429 nm and 558 nm. D) Difference spectra of HEK293 cells stably expressing NOX5ß grown in normal media (blue) or heme-depleted media with SA treatment (gray). NOX5 peaks are diminished in heme-depleted conditions. E) Difference spectra of HEK293 cells stably expressing NOX5ß grown in heme-depleted media with SA treatment with (red) and without (gray) a 2 h incubation with 5 µM heme. Heme addition increases the NOX5 specific peaks. F) Western blot analysis of cell lysates of the samples shown in E. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

Fig 3: Heme binding and its effect on NOX5 activity is dynamic. All cells were treated with cycloheximide before heme addition. A) Decay of the effect of heme on NOX5 activity over time. NOX5ß expressing cells were incubated with 1, 5 or 10 µM hemin (2 h). After incubation the media was removed and replaced with fresh media containing cycloheximide but no hemin and changes in NOX5 superoxide production was measured over time. Curves represent superoxide production normalized to untreated control at each timepoint. Each point represents mean ± SEM, n = 2–6 with *** representing p < 0.0001 for both treatment (heme) and time, two-way anova. B) Reduced minus oxidized difference spectra of live cells using the Olis Clarity. After addition of cycloheximide to halt protein synthesis, cells were treated with either DMSO control or 5 µM hemin (2 h). After incubation, media was removed from a subset of the hemin treated cells and replaced. The samples were normalized to live cell number and the UV–vis spectrum was taken before and after addition of sodium dithionite. The oxidized spectrum was then subtracted from the reduced spectrum. Characteristic NOX5 peaks are visible at 429 nm and 558 nm. C) Western blot analysis of cell lysates from B), HEK293 cells stably expressing NOX5ß grown in normal media and treated with either DMSO control, or 5 µM hemin with or without subsequent removal.

Fig 4: Heme addition restores superoxide production by NOX5 in cells grown in heme-depleted conditions, and increases superoxide production by cells grown in normal media. Superoxide levels were detected using the luminol analog L-012, which luminescences in the presence of superoxide but not other reactive oxygen species such as hydrogen peroxide. Before any incubation with hemin the cells were treated with cycloheximide to block protein translation. Addition of ionomycin Ca2+ salt triggers a NOX5 superoxide burst. A) Timecourse of superoxide production by HEK293 cells stably expressing NOX5ß grown in normal media vs. heme-depleted media with SA treatment. Addition of 5 µM hemin to cell culture for 2 h results in recovery of NOX5 superoxide production. B) Quantification of peak superoxide production by NOX5. Values represent means ± SEM, n = 10. Differences between columns assessed using a one-way anova with Tukey's test *p = 0.05, **p < 0.01. C) Timecourse of superoxide production by HEK293 cells stably expressing NOX5ß grown in heme-depleted media with SA treatment incubated with 5 µM hemin for 0, 10, or 120 min. D) Quantification of peak superoxide production by NOX5 with heme incubation. Values represent means ± SEM, n = 3. Differences between columns assessed using a one-way anova with Tukey's test **p < 0.01, ***p < 0.001. E) Peak superoxide production after incubation with hemin by HEK293 cells stably expressing NOX5ß in either normal media or heme-depleted media with SA treatment. Values represent means ± SEM, n = 6–12. Within each media group (normal media and heme-depleted with SA treatment) the differences in NOX5 activity with various heme concentrations was assessed using a one-way anova with Tukey's test with * denoting p < 0.05, **p < 0.01 and ***p < 0.001. F) Western blot analysis of cell lysates from HEK293 cells stably expressing NOX5ß grown in normal media and treated with either DMSO control or 5 µM hemin for 2 h.

Fig 5: Heme and radicicol have opposing effects on NOX5 activity and oligomerization. A) Cycloheximide was added to cells before hemin and/or radicicol treatment. Peak superoxide production by HEK293 cells stably expressing NOX5ß in either normal media or heme-depleted media with SA treatment after incubation with 0, 1, 5 or 10 µM hemin with or without 40 µM radicicol. Values are normalized to untreated control and represent means ± SEM, n = 3–12, *p = 0.05, **p < 0.01, ***p < 0.0001, two-tailed t-test. B) Western blot analysis of NOX5ß expression after 2 h incubation with 5 µM hemin with or without 40 µM radicicol. C) Reduced minus oxidized difference spectra of live cells using the Olis Clarity. After addition of cycloheximide to halt protein synthesis, HEK293 cells stably expressing NOXß grown in heme depleted media with SA were treated with vehicle control (gray), hemin (5 µM, 2 h, red) or hemin and radicicol (5 µM hemin, 40 µM radicicol, 2 h, dashed black). Cells were normalized to live cell number and the UV–vis spectrum was taken before and after addition of sodium dithionite. The oxidized spectrum was then subtracted from the reduced spectrum. Characteristic NOX5 peaks are visible at 429 nm and 558 nm. D, E) Co-immunoprecipitation of Hsp90 with NOX5. HEK293 cells stably expressing NOXß were treated with 40 µM radicicol and lysed. Equal protein amounts of lysate were incubated with NOX5 antibody and pulled down with Protein A Sepharose beads. After washing the precipitates were subjected to SDS-page and western blot analysis. D) HEK293 cells stably expressing NOXß were treated with 40 µM radicicol for 0, 0.5, 1 or 2 h. E) HEK293 cells stably expressing NOXß were treated with vehicle control or 40 µM radicicol and 1 µM ionomycin Ca2+ salt was added just prior to lysing. F) The interaction between Hsp90 and NOX5ß in HEK293 cells was probed using the Proximity Ligation Assay (PLA) Duolink kit from Sigma. Red foci represent interaction between the two proteins (<40 nm). Addition of only one antibody does not result in any signal, however, addition of both antibodies results in bright red foci indicating an interaction. HEK293 cells stably expressing NOXß were treated with vehicle control, 5 µM hemin, 40 µM radicicol, or both hemin and radicicol and interactions between NOX5 and Hsp90 were probed using the Duolink kit. G, H) HEK293 cells stably expressing NOXß were treated with vehicle control, 5 µM hemin or 40 µM radicicol for 2 h and then lysed. Samples were subjected to SDS-page and western blot analysis in semi-reducing conditions (sample preparation included no reducing agent and heating at 65 °C for 2 min) and probed for NOX5. G) Representative western blots and quantification of NOX5 oligomer/monomer in the G) absence or H) presence of 1 µM ionomycin Ca2+ salt. Values represent mean ± SEM, n = 3. Difference between treatments was assessed using a one-way anova with Tukey's test with * denoting p < 0.05 and **p < 0.01. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)