Fig 2: Purification of STRA6 from Pichia pastoris. (A) Fed-batch fermentation of recombinant Pichia pastoris expressing STRA6-GFP. Changes in the wet cell weight of yeast (red line) and dissolved oxygen (blue line) are shown over time. (B) Elution profile of Ni-NTA-purified STRA6-GFP on a Superdex 200, 10/300 GL column (blue trace) and after removal of GFP by treatment with HRV 3C protease (red trace). The disappearance of the peak corresponding to STRA6-GFP and appearance of a peak corresponding to GFP are clearly visible. The concentration of protein was determined by comparing the fluorescence of the eluted fractions to a standard curve of GFP and correcting for the molecular weight of the fusion protein. The void volume was determined using the elution profiles of blue dextran. (C) A silver-stained 10% SDS gel showing the purification of STRA6-GFP. The lanes are: lane 1, MW markers; lane 2, microsomes, 10 µg; lane 3, solubilized microsomes after extraction with C12E9, 10 µg; lane 4, flow-through from Ni-NTA column, 10 µg; lane 5, 250 mM imidazole eluate from Ni-NTA column, 3 µg; lane 6, eluate from Superdex 200, 10/300 GL column, 3 µg. (D) Immunoblot of STRA6-GFP after purification by Ni-NTA and size exclusion chromatography (lane 1) and following treatment with HRV 3C protease to remove the GFP (lane 2) using an antibody against full-length STRA6. The positions corresponding to the predicted molecular weights of STRA6-GFP and STRA6 are indicated.

Fig 3: Functional expression of STRA6-GFP in Pichia pastoris. (A) Schematic of the STRA6-GFP-c-Myc-His6 construct designed in this study. The location of a HRV 3C protease cleavage site is indicated by the red cross. (B) Co-localization of STRA6-GFP and DyLight594-conjugated holo-RBP at the cell surface of Pichia pastoris as determined by confocal microscopy. Top panel: GFP fluorescence, indicating the location of STRA6-GFP at the cell surface. Second panel: DyLight-594 fluorescence, showing the binding of RBP to the surface of cells transformed with STRA6-GFP but not empty vector. Third panel: merged image showing the colocalization of STRA6-GFP and DyLight-594 RBP at the cell surface. Bottom panel: Differential interference contrast (DIC) image of the yeast cells. Scale bar: 5 µm (C) Co-purification of holo-RBP with broken cells isolated from yeast transformed with STRA6-GFP but not with broken cells from empty-vector-transformed cells.

Fig 4: Interaction of apo-/holo-RBP with STRA6 determined by surface plasmon resonance (SPR).Representative SPR data for the binding of various concentrations of (A) apo-RBP and (B) holo-RBP to immobilized STRA6-GFP. Re-plot of triplicate independent steady-state responses at various concentrations of (C) apo-RBP and (D) holo-RBP against the total concentration of RBP added. The solid lines represent the total binding of RBP to the receptor as estimated from the solution to equation (1). The dashed lines represent the specific binding of RBP to the receptor without the contribution from non-specific binding. This component was derived by plotting only the specific component of the non-linear regression solution, i.e. (Rmax·[L]KD+[L]) from equation (1).

Fig 5: RT-qPCR data for BMECs treated with different ROH modalities for two hours.Expression values are normalized to the housekeeping gene, ACTB, and quantified relative to BMECs treated with HBSS alone. ??Cq data are presented in box-and-whisker format after log2 transformation with the values for each biological replicate displayed individually (N=4). Statistical analyses were performed in Prism on log2 transformed ??Cq values via one-way ANOVA followed by Dunnett’s test using a confidence interval of 95%. A) STRA6; B) CRBP1; C) LRAT.